JP2009006984A - Leg wheel type robot - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a leg wheel type robot which is suitable for reducing a driving torque at the time of turning by suppressing the increasing of the weight and the energy consumption amount. <P>SOLUTION: This leg wheel type robot 100 is equipped with a base body 10, a leg section 12 which is connected with the base body 10 while having a degree of freedom around the yawing axis, a driving wheel 20 which is rotatably installed on the leg section 12, and a joint motor 40 which gives power in order to drive the leg section 12 within the range of the degree of freedom around the yawing axis. The driving control of the leg section 12 at the time of an obstacle avoidance and the steering control of the driving wheel 20 at the time of running are performed by sharing the joint motor 40. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a leg-wheel type robot that can move over a step, and more particularly to a leg-wheel type robot that is suitable for suppressing an increase in weight and energy consumption and reducing driving torque during turning.

  The moving mechanism of the robot is classified into a wheel type, a crawler type, a leg type, or a combination thereof. In general, a wheel type robot has a problem of high mobility on a flat ground but low adaptability to a step. Crawler type robots are suitable for rough terrain, and can be overcome if there are a few steps. However, because they cannot actively move the center of gravity, they are less adaptable to steep stairs and are more mobile on flat ground than wheel type robots. There was a problem of being low. In addition, the legged robot has the best adaptability to the stairs, but has the problem of extremely low mobility on flat ground.

In order to solve the problems of the wheel type, the crawler type and the leg type, a leg wheel type robot combining the leg type and the wheel type has been proposed. As a leg wheel type robot, for example, a technique described in Patent Document 1 is known.
In Patent Document 1, walking movement is performed by freely driving four legs driven by a servo motor, and two legs provided on two rear legs of the four legs are provided. There is disclosed a legged mobile robot device that performs wheel movement by a driving wheel and two driven wheels provided on two front leg portions, and further hybrid movement that combines walking movement and wheel movement (same as above). Document [0123] and FIG. 31 (a)).

In each leg unit constituting the leg part, the upper link and the lower link are connected by a second connecting part driven by an actuator, and the upper link is connected to the body part by the first connecting part. In addition, a rotating part, which is a steering mechanism for rotating the driven wheel about the yaw axis, is provided between the first connecting part and the second connecting part of the front left leg unit and the right leg unit. Yes. Therefore, since the lower part of the front left and right leg unit is directed by the steering mechanism of the rotating unit, the legged mobile robot device can rotate and turn left or right (the same document [0124]- [0126]).
JP 2004-34169 A

By the way, as a steering mechanism at the time of travel of the leg-wheel type robot, a rotary joint that rotates around the yaw axis and an actuator that applies power for rotationally driving the rotary joint are provided at the leg tip, and steering control is performed using this actuator. A mechanism is conceivable.
However, since the rotary joint and the actuator are provided at the leg tip, both weight and energy consumption increase.

Further, even in the technique described in Patent Document 1, since a dedicated steering mechanism is provided for steering control at the time of traveling, there is a similar problem. There is also a problem that the driving torque of the steering mechanism increases due to the movement of the center of gravity of the steering mechanism.
Therefore, the present invention has been made paying attention to such an unsolved problem of the conventional technology, and is intended to suppress an increase in weight and energy consumption and to reduce driving torque during turning. An object is to provide a suitable leg-wheel type robot.

  [Invention 1] In order to achieve the above object, a leg-wheel type robot according to Invention 1 includes a base, a leg connected to the base with at least a degree of freedom around the yaw axis, and the leg. A wheel provided rotatably on the yaw axis, an actuator for applying power for driving the leg within a range of freedom around the yaw axis, and a control means for controlling the actuator. A leg-wheel type robot that moves by driving and rotation of the wheel, wherein the control means uses the actuator in common for driving control of the leg portion during obstacle avoidance and steering control of the wheel during traveling. Do it.

With such a configuration, in a place where an obstacle such as a flat ground does not exist, the steering control of the wheel is performed by the control means using the actuator while moving by running the wheel. When the actuator is driven, power is applied to the leg portion, the leg portion is driven within a range of the degree of freedom around the yaw axis, and the wheel is steered. Therefore, the mobility on the flat ground is high like the wheel type robot.
Further, where there is an obstacle such as a step, the control means performs drive control of the leg using an actuator. When the actuator is driven, power is applied to the leg portion, and the leg portion is driven within a range of degrees of freedom around the yaw axis, so that the step can be overcome. Therefore, the adaptability to a level difference is high like a legged robot.

  [Invention 2] Further, the leg-wheel type robot of Invention 2 includes a base and a leg connected to the base with a degree of freedom around the yaw axis and a degree of freedom around the pitch axis or roll axis, A wheel rotatably provided on the leg, a first actuator for applying power for driving the leg within a range of freedom around the yaw axis, and a freedom around the pitch axis or roll axis A second actuator for applying power for driving the leg within the range of the above, and a control means for controlling the first actuator and the second actuator, and is moved by driving the leg and rotating the wheel The leg wheel type robot, wherein the control means performs the drive control of the leg portion at the time of obstacle avoidance using the first actuator and the second actuator. , The steering control of the wheel during traveling is performed using the first actuator.

  With such a configuration, in a place where there is no obstacle such as a flat ground, the steering control of the wheel is performed by the control means using the first actuator while moving by running the wheel. When the actuator is driven, power is applied to the leg portion, the leg portion is driven within a range of the degree of freedom around the yaw axis, and the wheel is steered. Therefore, the mobility on the flat ground is high like the wheel type robot.

  Further, where there is an obstacle such as a step, the control means performs drive control of the leg using the first actuator and the second actuator. When the actuator is driven, power is applied to the leg portion, and the leg portion is driven within the range of the degree of freedom around the yaw axis and the range of the degree of freedom around the pitch axis or roll axis, and the step can be overcome. Therefore, the adaptability to a level difference is high like a legged robot.

  [Invention 3] Further, the leg-wheel type robot of Invention 3 is the leg-wheel type robot of Invention 2, comprising a plurality of the leg portions and a plurality of the wheels rotatably provided on the leg portions, The control means controls the first actuator and the second actuator so that a positional relationship between a polygon formed by a ground contact point of each wheel and a projected point of the center of gravity of the base body is constant during turning. To do.

  With such a configuration, at the time of turning, the control means causes the first actuator and the first actuator so that the positional relationship between the polygon formed by the contact points of the wheels and the projected point of the center of gravity of the base is constant. Two actuators are controlled.

[Invention 4] Further, the leg wheel type robot of the invention 4 is the leg wheel type robot of the invention 3, wherein the control means is configured such that, when turning, the center of the polygon formed by the ground contact points of the wheels and the base body. The first actuator and the second actuator are controlled so that the projected point of the center of gravity coincides.
With such a configuration, at the time of turning, the first actuator and the second actuator are controlled by the control means so that the center of the polygon formed by the ground contact point of each wheel coincides with the projected point of the center of gravity of the base. Is controlled.

[Invention 5] Further, the leg-wheel type robot according to Invention 5 is the leg-wheel type robot according to any one of Inventions 3 and 4, wherein the control means is configured such that, when turning, the wheel is a leg portion provided with the wheel. The first actuator and the second actuator are controlled so that the portion having the degree of freedom around the yaw axis is positioned on the axis.
With such a configuration, during turning, the control means causes the wheel to be positioned on the axis of the portion having the degree of freedom around the yaw axis in the leg portion provided with the wheel. The actuator and the second actuator are controlled.

[Invention 6] Further, the leg-wheel type robot of the invention 6 is the leg-wheel type robot according to any one of the inventions 2 to 5, wherein the control means displaces each of the wheels when the base is displaced in the vertical direction. The first actuator and the second actuator are controlled so that the positional relationship between the polygon formed by the ground contact point and the projected point of the center of gravity of the base is constant.
With such a configuration, when the base body is displaced in the vertical direction, the control means makes the positional relationship between the polygon formed by the contact points of the wheels and the projected point of the center of gravity of the base body constant. The first actuator and the second actuator are controlled.

[Invention 7] Furthermore, the leg-wheel type robot of the invention 7 is the leg-wheel type robot according to any one of the inventions 2 to 6, wherein the control means is a self-leg wheel type while maintaining the orientation of the base in a fixed direction. The first actuator is controlled so that the traveling direction of the robot matches the traveling direction of each wheel.
With this configuration, when the self-legged wheel robot is moving, the traveling direction of the self-legged robot matches the traveling direction of each wheel while maintaining the base of the leg-wheel robot at a fixed direction. Thus, the first actuator is controlled.

[Invention 8] Further, the leg-wheel robot according to Invention 8 is the leg-wheel robot according to any one of Inventions 2 to 7, wherein a plurality of the leg portions, the wheels, and the first actuator are provided with respect to the base. And a leg wheel mechanism unit configured by the second actuator, and the control means during the turning of the wheels when the base body rotates around the yaw axis at a predetermined rotation center position during turning. The first actuator is controlled so that the direction of movement of the center of rotation at the contact point between the arc trajectory drawn by the center of rotation and the center of rotation during steering coincides with the traveling direction of the wheels.
With such a configuration, during turning, the control means rotates the base around the yaw axis (vertical axis) and the arc trajectory drawn by the rotation center at the time of steering of each wheel and the rotation center at the time of steering. The first actuator is controlled so that the direction of movement of the center of rotation at the contact point with the direction of travel of each wheel coincides.

  [Invention 9] On the other hand, in order to achieve the above object, an object recognition apparatus according to Invention 9 is an object recognition apparatus for recognizing a surface or a boundary of a surface on an object, and measures a distance to a measurement point on the object. A distance measuring sensor, a scanning means for scanning the distance measuring sensor, a measurement result acquiring means for acquiring a measurement result of the distance measuring sensor for the measurement points that can be measured in a scanning range of the scanning means, and the measurement result A coordinate conversion means for converting the measurement result acquired by the acquisition means into coordinates of an orthogonal coordinate system; and a line segment in the orthogonal coordinate system by Hough conversion based on the coordinates of at least two of the measurement points converted by the coordinate conversion means. And a recognition unit for recognizing a surface on the object or a boundary of the surface based on the line segment detected by the line segment detection unit.

With such a configuration, the distance measuring sensor can be scanned by the scanning unit. Therefore, at least the planar shape of the object can be grasped. Then, the measurement result acquisition unit acquires the measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning unit, and the coordinate conversion unit converts the acquired measurement result into the coordinates of the orthogonal coordinate system. .
Next, a line segment in the orthogonal coordinate system is detected by the Hough transform based on the transformed coordinates of at least two measurement points.

As another method for detecting a line segment based on a plurality of measurement points, a least square method is known.
However, since the least square method detects a line segment by tracing each measurement point, a line segment that does not follow the surface on the object may be detected in a region where the measurement resolution is low. On the other hand, in the Hough transform, it is difficult to be affected by the measurement resolution, and a line segment relatively along the surface on the object can be detected even if the measurement resolution includes a low region.

  In the least-square method, line segments are detected by tracing each measurement point. Therefore, if measurement results vary due to the gloss of the measurement surface, etc. An area (hereinafter referred to as an error area) is detected as an oblique line segment, which is actually a flat line segment. On the other hand, the Hough transform is not easily affected by variations, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as flat line segments.

  Further, the least square method has a problem as to whether to recognize a region from where to where is one continuous surface. In this case, for example, it is conceivable to recognize a point where the slope of the detected line segment changes abruptly as the boundary of the continuous surface. However, with this recognition method, the error region and the regions on both sides of the error region are actually 1 Two continuous areas are recognized as separate continuous areas. On the other hand, the Hough transform is not easily affected by variation, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as one flat line segment.

When the line segment is detected by the Hough transform, the recognition unit recognizes the surface or the boundary of the surface on the object based on the detected line segment. Accordingly, it is possible to obtain a recognition result suitable for posture control of a robot that requires complicated posture control, such as a legged robot or a leg-wheel type robot.
Here, the scanning means may have any configuration as long as it scans the distance measuring sensor. For example, the scanning means scans the distance measuring sensor one-dimensionally so that the locus of the measurement points forms a line. Alternatively, the distance measuring sensor may be scanned two-dimensionally so that the locus of the measurement points forms a plane. In the former case, the planar shape of the object can be grasped, and in the latter case, the three-dimensional shape of the object can be grasped. The same applies to the object recognition apparatuses according to the tenth and fifteenth aspects.

Moreover, as a scanning means, the following structure is employable, for example.
(1) Rotating mechanism The rotating mechanism is configured by rotating means for rotating the distance measuring sensor around at least one scanning axis that forms a predetermined angle with respect to the measuring direction of the distance measuring sensor.
(2) Moving mechanism The moving mechanism is configured by moving means for moving the distance measuring sensor in at least one scanning direction different from the measuring direction of the distance measuring sensor. The moving means may move the distance measuring sensor along a path including a path extending in the scanning direction. The same applies to the object recognition apparatuses according to the tenth and fifteenth aspects.

  [Invention 10] The object recognition apparatus according to Invention 10 is an object recognition apparatus for recognizing a surface or a boundary of a surface on an object, the distance measuring sensor for measuring a distance to a measurement point on the object, and the measurement. Scanning means for scanning the distance sensor, measurement result acquiring means for acquiring the measurement results of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning means, and measurement results acquired by the measurement result acquiring means Coordinate conversion means for converting into coordinates in an orthogonal coordinate system, inter-measurement point interpolation means for interpolating between the measurement points converted by the coordinate conversion means, and points on the line obtained by the inter-measurement point interpolation means Line segment detection means for detecting a line segment in the orthogonal coordinate system by Hough transform based on the coordinates of the object, and recognition means for recognizing a surface or a boundary of the surface on the object based on the line segment detected by the line segment detection means With.

With such a configuration, the distance measuring sensor can be scanned by the scanning unit. Therefore, at least the planar shape of the object can be grasped. Then, the measurement result acquisition unit acquires the measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning unit, and the coordinate conversion unit converts the acquired measurement result into the coordinates of the orthogonal coordinate system. .
Next, the inter-measurement point interpolation means interpolates between the converted measurement points, and the line segment detection means detects the line segment in the orthogonal coordinate system by Hough transform based on the coordinates of the point on the obtained line. The

As another method for detecting a line segment based on a plurality of measurement points, a least square method is known.
However, since the least square method detects a line segment by tracing each measurement point, a line segment that does not follow the surface on the object may be detected in a region where the measurement resolution is low. On the other hand, in the Hough transform, it is difficult to be affected by the measurement resolution, and a line segment relatively along the surface on the object can be detected even if the measurement resolution includes a low region.

  In addition, in the least square method, the line segment is detected by tracing each measurement point. Therefore, when the measurement result varies due to the influence of the gloss of the measurement surface, the error area is actually a flat line segment. However, it is detected as an oblique line segment. On the other hand, the Hough transform is not easily affected by variations, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as flat line segments.

  Further, the least square method has a problem as to whether to recognize a region from where to where is one continuous surface. In this case, for example, it is conceivable to recognize a point where the slope of the detected line segment changes abruptly as the boundary of the continuous surface. However, with this recognition method, the error region and the regions on both sides of the error region are actually 1 Two continuous areas are recognized as separate continuous areas. On the other hand, the Hough transform is not easily affected by variation, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as one flat line segment.

When the line segment is detected by the Hough transform, the recognition unit recognizes the surface or the boundary of the surface on the object based on the detected line segment. Accordingly, it is possible to obtain a recognition result suitable for posture control of a robot that requires complicated posture control, such as a legged robot or a leg-wheel type robot.
Here, the interpolation includes connecting the measurement points with a line and approximating the measurement points with a line. The measurement points do not necessarily have to be located on the line, and adjacent measurement points are not necessarily included. It is not necessary to target points. The same applies to the object recognition device according to the seventeenth aspect and the object recognition method according to the twenty-sixth aspect.
The line includes a straight line, a line segment, a multi-order curve, and other curves. The same applies to the object recognition device according to the seventeenth aspect and the object recognition method according to the twenty-sixth aspect.

[Invention 11] The object recognition apparatus according to Invention 11 is the object recognition apparatus according to any one of Inventions 9 and 10, wherein the recognition means is based on the coordinates of the end points of the line segment detected by the line segment detection means. Recognize the boundary of the surface on the object.
With such a configuration, the recognition means recognizes the boundary of the surface on the object based on the coordinates of the end points of the detected line segment.

  [Invention 12] Further, the object recognition device of Invention 12 is the object recognition device of any one of Inventions 9 to 11, wherein the scanning means measures the measurement in a first scanning direction different from the measurement direction of the distance measuring sensor. A first scanning unit that scans the distance sensor; and a second scanning unit that scans the distance measuring sensor in a second scanning direction different from the measurement direction and the first scanning direction. The measurement result of the distance measuring sensor is acquired for the measurement points that can be measured in the scanning range of the first scanning means and the second scanning means.

  With this configuration, the first scanning unit can scan the distance measuring sensor in the first scanning direction, and the second scanning unit can scan the distance measuring sensor in the second scanning direction. Therefore, the three-dimensional shape of the object can be grasped. Then, the measurement result acquisition means acquires the measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the first scanning means and the second scanning means.

Here, as the first scanning unit and the second scanning unit, for example, the following configuration can be adopted.
(1) Rotation mechanism First rotation means for rotating the distance measuring sensor around a first scanning axis that forms a predetermined angle with respect to the measurement direction of the distance measuring sensor, and the measurement direction and the first scanning axis. And a second rotating means for rotating the distance measuring sensor around a second scanning axis forming a predetermined angle.
(2) Moving mechanism First moving means for moving the distance measuring sensor in a first scanning direction different from the measuring direction of the distance measuring sensor, and a second scanning direction different from the measuring direction and the first scanning direction. It comprises a second moving means for moving the distance measuring sensor. The first moving unit may move the distance measuring sensor along a first path including a path extending in the first scanning direction, and the second moving unit extends in the second scanning direction. The distance measuring sensor may be moved along a second route including a route to be performed. The same applies to (3) and (4) below.
(3) Combination of rotating mechanism and moving mechanism Rotating means for rotating the distance measuring sensor around a scanning axis forming a predetermined angle with respect to the measuring direction of the distance measuring sensor, and scanning different from the axial direction of the scanning axis And a moving means for moving the distance measuring sensor in the direction.
(4) A combination of a moving mechanism and a rotating mechanism The moving means for moving the distance measuring sensor in a scanning direction different from the measuring direction of the distance measuring sensor, and the scanning axis about a predetermined angle with respect to the scanning direction It is the structure which consists of a rotation means to rotate a ranging sensor. The same applies to the object recognition apparatus of the nineteenth aspect.

[Invention 13] The object recognition apparatus of Invention 13 is the object recognition apparatus of Invention 12, wherein the recognition means recognizes a surface on the object based on the coordinates of the end points of the line segment detected by the line segment detection means. To do.
With such a configuration, the surface on the object is recognized by the recognition unit based on the coordinates of the end points of the detected line segment.

  [Invention 14] The object recognition apparatus according to Invention 14 is the object recognition apparatus according to any one of Inventions 12 and 13, wherein the first scanning means has a first scanning axis that forms a predetermined angle with respect to the measurement direction. A first rotating means for rotating the distance measuring sensor around, the second scanning means being configured to rotate the distance measuring sensor around a second scanning axis that forms a predetermined angle with respect to the measuring direction and the first scanning axis; The measurement result acquisition means outputs the measurement result of the distance measurement sensor for each predetermined unit angle of the first rotation means while rotating the distance measurement sensor by the first rotation means. A predetermined unit angle of the first rotating means is obtained by performing a second scan in which the first scanning to be acquired is performed at predetermined unit angles of the second rotating means while rotating the distance measuring sensor by the second rotating means. Your And obtaining the measurement results for each predetermined unit angle of the second rotation means.

  With such a configuration, the distance measuring sensor can be rotated around the first scanning axis by the first rotating means, and the distance measuring sensor can be rotated around the second scanning axis by the second rotating means. Therefore, the three-dimensional shape of the object can be grasped. Then, the second scanning is performed by the measurement result acquisition unit, whereby measurement results for each predetermined unit angle of the first rotation unit and for each predetermined unit angle of the second rotation unit are acquired. In the second scanning, the first scanning is performed for each predetermined unit angle of the second rotating means while the distance measuring sensor is rotated by the second rotating means. In the first scan, a measurement result is acquired for each predetermined unit angle of the first rotating means while rotating the distance measuring sensor by the first rotating means.

  [Invention 15] The object recognition apparatus according to the invention 15 is an object recognition apparatus for recognizing a surface or boundary of a surface on an object, the distance measuring sensor for measuring a distance to a measurement point on the object, and the measurement A scanning unit that scans the distance sensor, a measurement result acquisition unit that acquires a measurement result of the distance measuring sensor for the measurement point that can be measured in a scanning range of the scanning unit, and a measurement result acquired by the measurement result acquisition unit. A first feature detecting unit for detecting a feature on the object based on the image, an imaging unit for capturing an image including the measurement point that can be measured in a scanning range of the scanning unit, and an object based on the image captured by the imaging unit. Second feature detection means for detecting the upper feature, and recognition means for recognizing a surface or a boundary of the surface on the object based on detection results of the first feature detection means and the second feature detection means.

With such a configuration, the distance measuring sensor can be scanned by the scanning unit. Therefore, at least the planar shape of the object can be grasped. Then, the measurement result acquisition means acquires the measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning means, and the first feature detection means determines the feature on the object based on the acquired measurement result. Detected.
In addition, an image including a measurement point that can be measured within the scanning range of the scanning unit is captured by the imaging unit, and a feature on the object is detected based on the captured image by the second feature detection unit.

Then, the recognition unit recognizes the surface or the boundary of the surface on the object based on the detection results of the first feature detection unit and the second feature detection unit.
Here, the features on the object include, for example, points indicating features on the object, straight lines, line segments, multi-order curves, circles, ellipses and other curves, planes, curved surfaces and other surfaces, cubes, spheres and other three-dimensional shapes. Features are included. The same applies to the object recognition method of the invention 27 below.

  [Invention 16] Further, the object recognition device of Invention 16 is the object recognition device of Invention 15, wherein the first feature detection means is a coordinate for converting the measurement result acquired by the measurement result acquisition means into coordinates of an orthogonal coordinate system. Conversion means; and line segment detection means for detecting line segments in the orthogonal coordinate system by Hough transform based on the coordinates of at least two measurement points converted by the coordinate conversion means; and the recognition means, Based on the line segment detected by the line segment detection means and the detection result of the second feature detection means, the surface on the object or the boundary of the surface is recognized.

If it is such a structure, the acquired measurement result will be converted into the coordinate of a rectangular coordinate system by a coordinate conversion means.
Then, a line segment in the orthogonal coordinate system is detected by the Hough transform based on the transformed coordinates of at least two measurement points.
As another method for detecting a line segment based on a plurality of measurement points, a least square method is known.

However, since the least square method detects a line segment by tracing each measurement point, a line segment that does not follow the surface on the object may be detected in a region where the measurement resolution is low. On the other hand, in the Hough transform, it is difficult to be affected by the measurement resolution, and a line segment relatively along the surface on the object can be detected even if the measurement resolution includes a low region.
In addition, in the least square method, the line segment is detected by tracing each measurement point. Therefore, when the measurement result varies due to the influence of the gloss of the measurement surface, the error area is actually a flat line segment. However, it is detected as an oblique line segment. On the other hand, the Hough transform is not easily affected by variations, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as flat line segments.

  Further, the least square method has a problem as to whether to recognize a region from where to where is one continuous surface. In this case, for example, it is conceivable to recognize a point where the slope of the detected line segment changes abruptly as the boundary of the continuous surface. However, with this recognition method, the error region and the regions on both sides of the error region are actually 1 Two continuous areas are recognized as separate continuous areas. On the other hand, the Hough transform is not easily affected by variation, and if the number of variations is small, the error region and the regions on both sides thereof can be detected as one flat line segment.

  When the line segment is detected by the Hough transform, the recognition unit recognizes the surface or the boundary of the surface on the object based on the detected line segment and the detection result of the second feature detection unit. Therefore, the technique of Non-Patent Document 1 can generate only simple map information for movement control of a crawler type robot that does not require complicated posture control. Since at least the planar shape of an object can be grasped as a boundary, a recognition result suitable for posture control of a robot that requires complicated posture control such as a legged robot or a leg-wheel type robot can be obtained. .

  [Invention 17] Further, the object recognition device of Invention 17 is the object recognition device of Invention 15, wherein the first feature detection means is a coordinate for converting the measurement result acquired by the measurement result acquisition means into coordinates of an orthogonal coordinate system. A conversion means, an inter-measurement point interpolation means for interpolating between the measurement points converted by the coordinate conversion means, and the Hough transform based on the coordinates of the points on the line obtained by the inter-measurement point interpolation means. A line segment detecting means for detecting a line segment in an orthogonal coordinate system, wherein the recognizing means is a surface on the object based on the line segment detected by the line segment detecting means and the detection result of the second feature detecting means. Or recognize the boundary of the face.

If it is such a structure, the acquired measurement result will be converted into the coordinate of a rectangular coordinate system by a coordinate conversion means, and between the measured points will be interpolated by a line by the inter-measurement point interpolation means.
Then, a line segment in the orthogonal coordinate system is detected by the Hough transform based on the coordinates of the obtained point on the line. It should be noted that there are advantages similar to those described above over the least square method.

  When the line segment is detected by the Hough transform, the recognition unit recognizes the surface or the boundary of the surface on the object based on the detected line segment and the detection result of the second feature detection unit. Therefore, as described above, it is possible to obtain a recognition result suitable for posture control of a robot that requires complicated posture control, such as a legged robot or a leg-wheel type robot.

  [Invention 18] Furthermore, the object recognition device of Invention 18 is the object recognition device of Invention 15, wherein the first feature detection means is a coordinate for converting the measurement result acquired by the measurement result acquisition means into coordinates of an orthogonal coordinate system. Inclination calculation for calculating the inclination at the measurement point in the orthogonal coordinate system based on the coordinates of the measurement point and the coordinates of the measurement points in the vicinity thereof for each measurement point converted by the conversion means and the coordinate conversion means And an appearance frequency calculating means for calculating the appearance frequency of each inclination with respect to the total number of inclinations calculated by the inclination calculating means, and the recognition means is the appearance frequency calculated by the appearance frequency calculating means and the converted The surface on the object or the boundary of the surface is recognized based on the coordinates of each measurement point and the detection result of the second feature detection means.

  If it is such a structure, the acquired measurement result will be converted into the coordinate of a rectangular coordinate system by the coordinate conversion means, and the coordinates of the measurement point and its surroundings will be obtained for each converted measurement point by the inclination calculation means. The inclination at the measurement point in the orthogonal coordinate system is calculated based on the coordinates of the measurement points, and the appearance frequency calculating means calculates the appearance frequency of each inclination with respect to the calculated total number of inclinations.

  When the appearance frequency of each inclination is calculated, the recognition unit calculates the surface on the object or the boundary of the surface based on the calculated appearance frequency, the converted coordinates of each measurement point, and the detection result of the second feature detection unit. Is recognized. Therefore, as described above, it is possible to obtain a recognition result suitable for posture control of a robot that requires complicated posture control, such as a legged robot or a leg-wheel type robot.

  Here, the inclination calculation means, for example, approximates a regression line with respect to a plurality of measurement points by a known least square method, calculates an inclination from the regression line, or, in an orthogonal coordinate system, a plurality of continuous in one axial direction. Inclination is calculated using the difference value of the coordinates of the two measurement points at both ends of the measurement point and the difference value of the coordinates of the measurement points of the two ends of the plurality of measurement points continuous in the other axial direction. Can do.

  [Invention 19] The object recognition apparatus according to Invention 19 is the object recognition apparatus according to any one of Inventions 15 to 18, wherein the scanning means measures the measurement in a first scanning direction different from the measurement direction of the distance measuring sensor. A first scanning unit that scans the distance sensor; and a second scanning unit that scans the distance measuring sensor in a second scanning direction different from the measurement direction and the first scanning direction. The measurement result of the distance measuring sensor is acquired for the measurement points that can be measured in the scanning range of the first scanning means and the second scanning means.

  With this configuration, the first scanning unit can scan the distance measuring sensor in the first scanning direction, and the second scanning unit can scan the distance measuring sensor in the second scanning direction. Therefore, the three-dimensional shape of the object can be grasped. Then, the measurement result acquisition means acquires the measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the first scanning means and the second scanning means.

  [Invention 20] Further, the object recognition device of Invention 20 is the object recognition device of Invention 19, wherein the first scanning means is the distance measuring sensor around a first scanning axis that forms a predetermined angle with respect to the measurement direction. The second scanning means rotates the distance measuring sensor around a second scanning axis that forms a predetermined angle with respect to the measurement direction and the first scanning axis. The measurement result acquisition means performs a first scan for acquiring the measurement result of the distance measurement sensor for each predetermined unit angle of the first rotation means while rotating the distance measurement sensor by the first rotation means. By performing the second scanning performed at every predetermined unit angle of the second rotating means while rotating the distance measuring sensor by the second rotating means, the second rotation is performed at every predetermined unit angle of the first rotating means. hand Acquiring the measurement results for each predetermined unit angle.

  With such a configuration, the distance measuring sensor can be rotated around the first scanning axis by the first rotating means, and the distance measuring sensor can be rotated around the second scanning axis by the second rotating means. Therefore, the three-dimensional shape of the object can be grasped. Then, the second scanning is performed by the measurement result acquisition unit, whereby measurement results for each predetermined unit angle of the first rotation unit and for each predetermined unit angle of the second rotation unit are acquired. In the second scanning, the first scanning is performed for each predetermined unit angle of the second rotating means while the distance measuring sensor is rotated by the second rotating means. In the first scan, a measurement result is acquired for each predetermined unit angle of the first rotating means while rotating the distance measuring sensor by the first rotating means.

  [Invention 21] Further, the object recognition device of Invention 21 is the object recognition device of any one of Inventions 15 to 18, comprising a plurality of the distance measuring sensors and the scanning means, and each of the scanning means is the scanning means. Are scanned in a first scanning direction different from the measurement direction of the distance measuring sensor, and each distance measuring sensor is scanned in a second scanning direction different from the measurement direction and the first scanning direction. Is arranged.

  With such a configuration, each scanning unit can scan each distance measuring sensor arranged in the second scanning direction in the first scanning direction. Therefore, the three-dimensional shape of the object can be grasped. And the measurement result of each ranging sensor is acquired by the measurement result acquisition means about the measurement point which can be measured in the scanning range of a scanning means.

  [Invention 22] The object recognition apparatus according to Invention 22 is the object recognition apparatus according to any one of Inventions 15 to 21, wherein the first feature detection means is an object based on the measurement result acquired by the measurement result acquisition means. The upper feature point is detected, and the second feature detection unit includes an image line segment detection unit that detects a line segment from an image captured by the imaging unit, and the recognition unit is based on the imaging unit. In the coordinate system, the surface on the object or the boundary of the surface is recognized based on the positional relationship between the line segment detected by the image line segment detection unit and the feature point detected by the first feature detection unit.

If it is such a structure, the feature point on an object will be detected by the 1st feature detection means based on the acquired measurement result, and a line segment will be detected from the imaged image by the image line segment detection means. .
Then, the recognition unit recognizes the surface or the boundary of the surface on the object based on the positional relationship between the detected line segment and the detected feature point in the coordinate system based on the imaging unit.

  [Invention 23] Furthermore, the object recognition device of Invention 23 is the object recognition device of Invention 22, wherein the recognition means calculates the line segment and the line segment for each line segment detected by the image line segment detection means. Corresponding to the feature points existing within a predetermined distance from the extended straight line, grouping the line segments associated with the same feature points, and based on the coordinates of the end points of the line segments belonging to the same group Recognize the upper face or face boundary.

  With such a configuration, the recognition means associates each detected line segment with the line segment and a feature point existing within a predetermined distance from the straight line obtained by extending the line segment, and the same feature The line segments associated with the points are grouped, and the surface on the object or the boundary of the surface is recognized based on the coordinates of the end points of the line segments belonging to the same group.

  [Invention 24] Further, the object recognition device according to Invention 24 is the object recognition device according to any one of Inventions 22 and 23, wherein a plane composed of two predetermined axes in an orthogonal coordinate system based on the distance measuring sensor, The distance measuring sensor and the imaging means are arranged so that a plane composed of two predetermined axes in an orthogonal coordinate system based on the imaging means is parallel.

  [Invention 25] On the other hand, in order to achieve the above object, the object recognition method of Invention 25 recognizes a surface or a boundary of a surface on an object using a distance measuring sensor that measures a distance to a measurement point on the object. An object recognition method, a scanning step of scanning the distance measuring sensor, a measurement result acquiring step of acquiring a measurement result of the distance measuring sensor for the measurement points that can be measured in a scanning range of the scanning step, and the measurement A coordinate conversion step for converting the measurement result acquired in the result acquisition step into a coordinate in the orthogonal coordinate system, and a line in the orthogonal coordinate system by Hough conversion based on the coordinates of the at least two measurement points converted in the coordinate conversion step. A line segment detecting step for detecting a segment, and a recognition step for recognizing a surface or a boundary of the surface on the object based on the line segment detected in the line segment detecting step.

  Here, the scanning step may be any method as long as it scans the distance measuring sensor. For example, the distance measuring sensor may be scanned one-dimensionally so that the locus of the measurement points forms a line. Alternatively, the distance measuring sensor may be scanned two-dimensionally so that the locus of the measurement points forms a plane. In the former case, the planar shape of the object can be grasped, and in the latter case, the three-dimensional shape of the object can be grasped. The same applies to the object recognition methods of inventions 26 and 27 below.

  [Invention 26] Further, the object recognition method of the invention 26 is an object recognition method for recognizing a surface or a boundary of a surface on an object by using a distance measuring sensor for measuring a distance to a measurement point on the object. A scanning step for scanning the distance measuring sensor, a measurement result acquiring step for acquiring a measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning step, and a measurement result acquired in the measurement result acquiring step On the line obtained by the coordinate conversion step for converting the coordinates into the coordinates of the orthogonal coordinate system, the inter-measurement point interpolation step for interpolating between the measurement points converted by the coordinate conversion step, and the inter-measurement point interpolation step A line segment detection step for detecting a line segment in the orthogonal coordinate system by Hough transform based on the coordinates of the point, and a surface on the object based on the line segment detected in the line segment detection step. And a recognition step recognizes the boundary of the surface.

  [Invention 27] The object recognition method of the invention 27 is an object recognition method for recognizing a surface or a boundary of a surface on an object by using a distance measuring sensor for measuring a distance to a measurement point on the object. A scanning step for scanning the distance measuring sensor, a measurement result acquiring step for acquiring a measurement result of the distance measuring sensor for the measurement points that can be measured in the scanning range of the scanning step, and a measurement result acquired in the measurement result acquiring step A first feature detection step for detecting a feature on the object based on the image, an imaging step for capturing an image including the measurement point that can be measured in a scanning range of the scanning step, and an image captured in the imaging step A second feature detecting step for detecting a feature on the object; a surface on the object based on the detection results of the first feature detecting step and the second feature detecting step; The boundary surface and a recognizing step.

  As described above, according to the leg-wheel type robot of the first aspect of the invention, since the leg drive control at the time of obstacle avoidance and the wheel steering control at the time of running are performed using the actuator in common, the steering at the time of running is performed. There is no need to provide a dedicated actuator on the leg for control, and an effect that an increase in weight and energy consumption can be suppressed as compared with the conventional case is obtained.

Furthermore, according to the leg-wheel type robot of the invention 2, the drive control of the leg part at the time of obstacle avoidance and the steering control of the wheel at the time of running are performed by using the first actuator in common. Therefore, there is no need to provide a dedicated actuator on the leg portion, and an effect that an increase in weight and energy consumption can be suppressed as compared with the conventional case can be obtained.
Furthermore, according to the leg-wheel type robot of the invention 3 or 4, at the time of turning, the positional relationship between the polygon formed by the ground contact point of each wheel and the projected point of the center of gravity of the base is constant, or each wheel Since the center of the polygon composed of the ground contact points and the projected point of the center of gravity of the base are matched, the fear of falling can be reduced by bringing the projected point of the center of gravity close enough to the center of the polygon. The effect is obtained.

Further, according to the leg-wheel type robot of the fifth aspect, since the length between the ground contact point of each wheel and the rotary joint is zero at the time of turning, there is an effect that the driving torque of the first actuator can be reduced. It is done.
Furthermore, according to the leg-wheel type robot of the sixth aspect, the base body can be moved up and down while maintaining the positional relationship between the polygon formed by the contact points of the wheels and the projected point of the center of gravity of the base body. Even in areas where the ceiling is limited, the effect is that activities can be performed without compromising the existing effects.

  Furthermore, according to the leg-wheel type robot of the invention 7, the leg-wheel type robot can be moved in a free direction regardless of the direction thereof, so that a quick movement in each direction can be realized. It is possible to perform activities even in areas where it is difficult to change the direction of the leg wheel type robot, such as a narrow and intricate area where each component of the leg wheel type robot is hindered and cannot turn. .

  Furthermore, according to the leg-wheel robot of the invention 8, at the time of turning, the arc drawn by the rotation center at the time of steering each wheel when the base of the leg-wheel robot is rotated around the yaw axis at a predetermined rotation center position. Since the direction of movement of the center of rotation at the contact point between the track and the center of rotation at the time of steering can coincide with the traveling direction of each wheel, the leg wheel type robot can be moved back and forth at a predetermined turning center position. The effect that a turn (a turn equivalent to a super-trust turn by a crawler or the like) can be performed is obtained. Further, by setting the center position of the base body as the rotation center position when rotating, the leg wheel type robot can be turned with the minimum turning radius.

  On the other hand, according to the object recognition device of the invention 9, 10, 16 or 17, it is possible to grasp at least a planar shape of an object as a surface on the object or a boundary of the surface. As described above, it is possible to obtain a recognition result suitable for posture control of a robot that requires complicated posture control. In addition, because line segments are detected by Hough transform, when object recognition is performed using a two-dimensional distance measuring device using a distance measuring sensor, the possibility that the recognition accuracy will decrease due to a decrease in measurement resolution or variations in measurement results is reduced. The effect that it can do is also acquired.

Further, according to the object recognition device of the invention 10 or 17, since the line segment is detected based on the coordinates of the points on the line obtained by interpolating between the measurement points, the measurement resolution is lowered or the measurement results are varied. Even if this occurs, it is possible to obtain a relatively accurate recognition result, and to further reduce the possibility that the recognition accuracy is lowered.
Furthermore, according to the object recognition device of the eleventh aspect of the present invention, the effect that the boundary of the surface on the object can be recognized relatively accurately is obtained.

Furthermore, according to the object recognition device of the twelfth aspect, since the three-dimensional shape of the object can be grasped as a surface on the object or a boundary between the surfaces, complicated posture control such as a legged robot or a leg wheel type robot can be performed. It is possible to obtain a recognition result more suitable for posture control of a robot that requires
Furthermore, according to the object recognition device of the thirteenth aspect, it is possible to recognize the surface on the object relatively accurately.

Furthermore, according to the object recognition device of the invention 14, since the rotation mechanism for rotating the distance measuring sensor is adopted, the space required for scanning is smaller than that of the moving mechanism, and the mechanism for scanning becomes simple. And the effect that a high-speed scanning is realizable is acquired.
Furthermore, according to the object recognition device of the fifteenth aspect, in addition to the distance measurement sensor, the feature on the object is detected using imaging means of a different system, and the surface on the object is detected based on the two detection results. Alternatively, since the boundary of the surface is recognized, the shortcomings of the method using the distance measuring sensor can be compensated for by the method using the imaging means, and the recognition accuracy decreases when the object is recognized by the two-dimensional distance measuring device using the distance measuring sensor. The effect that it is possible to reduce the possibility of doing is obtained. In addition, since it is not necessary to increase the number of scans of the scanning means, the effect that the measurement time can be shortened is also obtained.

  Furthermore, according to the object recognition device of the eighteenth aspect of the invention, since at least a planar shape of an object can be grasped as a surface on the object or a boundary between surfaces, a complicated posture such as a legged robot or a leg-wheel type robot can be obtained. An effect is obtained that a recognition result suitable for posture control of a robot that requires control can be obtained. In addition, since the surface on the object or the boundary of the surface is recognized based on the appearance frequency of the inclination at the measurement point and the coordinates of each measurement point, when the object recognition is performed by the two-dimensional distance measuring device using the distance measuring sensor. In addition, there is an effect that the possibility that the recognition accuracy is lowered can be further reduced.

Furthermore, according to the object recognition device of the invention 19 or 21, the three-dimensional shape of the object can be grasped as a surface on the object or a boundary between the surfaces. An effect is obtained that a recognition result more suitable for posture control of a robot that requires posture control can be obtained.
Furthermore, according to the object recognition device of the twentieth aspect, since the rotation mechanism that rotates the distance measuring sensor is adopted, the space required for scanning is smaller than that of the moving mechanism, and the mechanism for scanning becomes simple. And the effect that a high-speed scanning is realizable is acquired.

Furthermore, according to the object recognition device of the invention 22, the surface on the object or the boundary of the surface is recognized based on the positional relationship between the line segment detected from the image and the feature point detected from the measurement result of the distance measuring sensor. Therefore, an effect that the possibility that the recognition accuracy is lowered can be further reduced.
Further, according to the object recognition device of the invention 23, the distance measurement sensor measures even if it is detected as a plurality of line segments from the captured image due to the influence of contrast or the like, which should actually be detected as one line segment. Based on the correspondence with the feature points detected from the results, multiple line segments are grouped as a single line segment and the surface or boundary of the surface on the object is recognized, further reducing the possibility of reduced recognition accuracy The effect that it can do is acquired.

On the other hand, according to the object recognition method of the invention 25, an effect equivalent to that of the object recognition device of the invention 9 can be obtained.
Furthermore, according to the object recognition method of the invention 26, an effect equivalent to that of the object recognition device of the invention 10 can be obtained.
Furthermore, according to the object recognition method of the twenty-seventh aspect, an effect equivalent to that of the object recognition device of the fifteenth aspect can be obtained.

[First Embodiment]
Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. 1 to 10 are views showing a first embodiment of a leg-wheel type robot according to the present invention.
First, the configuration of a leg wheel type robot 100 to which the present invention is applied will be described.
FIG. 1 is a front view of a leg wheel type robot 100.

FIG. 2 is a side view of the leg wheel type robot 100.
As shown in FIGS. 1 and 2, the leg-wheel type robot 100 includes a base body 10 and four leg portions 12 coupled to the base body 10.
In front of the base 10, two leg portions 12 are coupled to a symmetrical position via a rotary joint 14. In addition, two legs 12 are connected to the rear side of the base body 10 via a rotary joint 14 at a symmetrical position. The rotary joint 14 rotates with the direction orthogonal to the bottom surface of the leg wheel type robot 100 as an axial direction. That is, it rotates around the yaw axis.

  Each leg portion 12 is provided with two rotary joints 16 and 18. The rotary joint 14 rotates with the lower side as the axial direction, and the rotary joints 16 and 18 rotate with the direction orthogonal to the side surface of the leg wheel type robot 100 as the axial direction when the rotary joint 14 is in the state of FIG. . That is, when the rotary joint 14 is in the state of FIG. 1, it rotates about the pitch axis, and when the rotary joint 14 is rotated 90 degrees from the state of FIG. 1, it rotates about the roll axis. Therefore, each leg 12 has three degrees of freedom.

A driving wheel 20 is rotatably provided at the tip of each leg 12 with the same axial direction as the rotary joints 16 and 18.
At the tip of each leg 12, a front leg tip sensor 22 for measuring the distance to an object existing on the movement path of the leg wheel type robot 100 and a lower leg tip sensor 24 for measuring the distance to the ground plane are provided. Is provided.

On the other hand, a horizontal laser 26 that irradiates a horizontal laser beam is provided in the upper center of the front surface of the substrate 10. In addition, vertical lasers 28 and 30 for irradiating vertical surface laser light are respectively provided on the center left and right of the front surface of the substrate 10.
A camera 32 that captures an image including reflected light of a horizontal plane laser beam and a vertical plane laser beam is provided at the lower center of the front surface of the substrate 10.

The horizontal laser 26 is provided so as to be inclined downward by a predetermined angle so that the camera 32 can capture an image including the reflected light of the horizontal laser beam. Similarly, the vertical laser 28 is tilted to the right by a predetermined angle so that the camera 32 can capture an image including reflected light of the vertical plane laser beam, and the vertical laser 30 is tilted to the left by a predetermined angle. ing.
Obstacle sensors 34 and 36 for detecting an obstacle are provided on the left and right of the camera 32, respectively.

FIG. 3 is a diagram illustrating the configuration of the obstacle sensors 34 and 36.
As shown in FIG. 3A, the obstacle sensors 34 and 36 can be configured by arranging a plurality of ultrasonic ranging sensors having low directivity in an array. Further, as shown in FIG. 3B, a plurality of infrared ranging sensors having high directivity can be arranged in an array. It is not limited to the configuration arranged in an array, and may be configured as a single unit. Moreover, you may comprise with the area sensor which arranged the ultrasonic ranging sensor or the infrared ranging sensor on the several plane. Thereby, the object which exists on the movement path | route of the leg wheel type robot 100 can be detected roughly.

Next, the movement control system of the leg wheel type robot 100 will be described.
FIG. 4 is a block diagram showing a movement control system of the leg wheel type robot 100.
As shown in FIG. 4, joint motors 40 that rotationally drive the rotary joints 14 to 18 are respectively provided in the rotary joints 14 to 18 of the respective leg portions 12. Each joint motor 40 is provided with an encoder 42 that detects the rotational angle position of the joint motor 40, and a driver 44 that controls the driving of the joint motor 40 based on the motor command signal and the output signal of the encoder 42.

A wheel motor 50 that rotationally drives the drive wheel 20 is provided on the drive wheel 20 of each leg 12. Each wheel motor 50 is provided with an encoder 52 that detects the rotational angle position of the wheel motor 50, and a driver 54 that controls the driving of the wheel motor 50 based on the motor command signal and the output signal of the encoder 52.
The leg wheel type robot 100 further performs wireless communication with the CPU 60, the three-axis attitude sensor 70 that detects the attitude of the leg wheel type robot 100, the vision processor 72 that processes the image signal of the camera 32, and an external PC. The wireless communication unit 74 is configured to include a vision processor 72, a wireless communication unit 74, a hub 76 that relays input / output of the CPU 60, and a speaker 78 that outputs a warning sound or the like.

The triaxial attitude sensor 70 includes a gyroscope or an acceleration sensor, or both, and detects the inclination of the attitude of the leg wheel type robot 100 with respect to the ground axis.
The CPU 60 outputs motor command signals to the drivers 44 and 54 via the motor command output I / F 61 and inputs output signals of the encoders 42 and 52 via the angle fetch I / F 62. In addition, sensor signals are input from the front leg tip sensor 22, the lower leg tip sensor 24, the obstacle sensor 34, and the triaxial posture sensor 70 via the sensor input I / F 63, respectively. Further, signals are input / output to / from the hub 76 via the communication I / F 64, and an audio signal is output to the speaker 78 via the sound output I / F 65.

Next, processing executed by the CPU 60 will be described.
The CPU 60 activates a control program stored in a predetermined area such as a ROM, and executes the travel control process and the elevation control process shown in the flowchart of FIG. 6 in a time-sharing manner according to the control program.
First, control processing during traveling will be described.

FIG. 5 is a diagram illustrating a running state of the leg wheel type robot 100.
Steering control of the leg wheel type robot 100 is performed so that the driving wheel 20 is positioned on the axis of the rotary joint 14 provided with the driving wheel 20 (hereinafter referred to as grounding on the yaw axis). This is performed by controlling the joint motor 40, and the left and right wheel steering angles θl and θr and the left and right wheel rotation angular velocities ωl and ωr are calculated by inputting the robot traveling direction speed Vc and the turning radius R of the robot center every moment. In addition, a command is given to each actuator.

  When the leg-wheel type robot 100 performs steering, the distance to each drive wheel 20 is formed concentrically with the center point of the turning radius R at the center of the robot in order to minimize the slip of each drive wheel 20. Therefore, the left and right wheel steering angles θl and θr can be calculated by the following equations (1) and (2), where Wb is the wheel base and Wt is the tread.

  Since each wheel is a drive wheel, in order to maintain the ground contact state on the yaw axis, the tangential speeds Vl and Vr of the wheel part must be given by the following equations (3) and (5). When the wheel diameter is D, the left and right wheel rotational angular velocities ωl and ωr can be calculated by the following equations (4) and (6).

  When the radius of rotation R is infinite, it can be approximated to straight movement, so there is no need to distinguish between straight and turning. Therefore, the operation transition from straight to turning and straight movement can be performed smoothly and continuously. It can be carried out.

Next, the elevation control process will be described.
FIG. 6 is a flowchart showing the elevation control process.
The elevation control process is a process for performing the elevation control of the leg portion 12. When the elevation control process is executed by the CPU 60, the process first proceeds to step S200 as shown in FIG.
In step S200, an image is captured from the vision processor 72, and the process proceeds to step S202.

In step S202, feature points of stairs are extracted by a light cutting method based on the captured image.
FIG. 7 is a diagram for explaining the principle of the light cutting method.
The light section method is a measurement method for obtaining coordinates on a measurement object based on the principle of triangulation. FIG. 7 shows the measurement coordinate system.

  The coordinates P (x0, y0, z0) on the measurement object are obtained by the following equation (7), where Ps (xi, yi, zi) is an arbitrary coordinate on the image sensor of the camera 32.

Next, from the obtained three-dimensional coordinates, a discontinuous point or a bent point of the reflected light of the laser beam is extracted as a feature point of the staircase.
FIG. 8 is a diagram illustrating a state in which the laser beam is irradiated on the stairs and an image of the imaging element of the camera 32.

  When a stairway is present on the movement path of the leg-wheel type robot 100, as shown on the left side of FIG. 8A, the horizontal surface laser light emitted from the horizontal laser 26 is reflected by the kick plate and the floor surface of the staircase, and the camera 32 Thus, an image of the stairs including the reflected light is taken. When image processing is performed on the image, the reflected light edge on the kick plate and the reflected light edge on the floor surface can be extracted as shown on the right side of FIG. Based on the edge image and the three-dimensional coordinates obtained by the above equation (7), the actual coordinates corresponding to the discontinuous points of the reflected light edge can be calculated.

  Further, as shown on the left side of FIG. 8B, the vertical plane laser light emitted from the vertical laser 28 is reflected by the stair kick plate and the step board, and the camera 32 captures an image of the stair including the reflected light. The When image processing is performed on the image, as shown on the right side of FIG. 8B, the reflected light edge on the kick board and the reflected light edge on the tread board can be extracted. The same applies to the vertical laser 30. As shown on the right side of FIG. 8C, the reflected light edge on the kick board and the reflected light edge on the tread board can be extracted. Based on these edge images and the three-dimensional coordinates obtained by the above equation (7), the actual coordinates for the bent point of the reflected light edge can be calculated.

Returning to FIG. 6, the process proceeds to step S <b> 204, and the width of the staircase is calculated based on the extracted feature points. The process proceeds to step S <b> 206, and the actual coordinates of the stair nosing part of the staircase are calculated based on the extracted feature points. Then, the process proceeds to step S208.
In step S208, inverse kinematics calculation and centroid calculation are performed based on the calculated width of the staircase and the actual coordinates of the nose and the sensor signal of the three-axis posture sensor 70, and the process proceeds to step S210 and the calculation result of step S208. The landing position of the leg tip (drive wheel 20) is determined based on the above, and the process proceeds to step S212.

  In step S212, sensor signals are input from the front leg tip sensor 22 and the lower leg tip sensor 24, respectively, the process proceeds to step S214, and the distance to the kick plate is calculated based on the input sensor signal of the front leg tip sensor 22. Then, the process proceeds to step S216, where the positional relationship between the leg tip and the tread is calculated based on the input sensor signal of the lower leg tip sensor 24, and the process proceeds to step S218.

In step S218, a motor command signal to the drivers 44 and 54 is generated based on the determined landing position and the calculated both distances, the process proceeds to step S220, and the generated motor command signal is output to the drivers 44 and 54. The process proceeds to step S222.
In step S222, it is determined whether or not the leg tip has landed on the tread. If it is determined that the leg tip has landed (Yes), the series of processes is terminated and the original process is restored.

On the other hand, when it is determined in step S222 that the leg tip does not land (No), the process proceeds to step S212.
Next, the operation of the present embodiment will be described.
On a flat ground, the leg wheel type robot 100 can move by wheel running. At the time of turning, the joint motor 40 is controlled so that each drive wheel 20 is grounded on the yaw axis.

FIG. 9 is a diagram illustrating an operation when each drive wheel 20 turns without grounding on the yaw axis.
FIG. 10 is a diagram showing an operation when each drive wheel 20 turns on the yaw axis while turning. In both figures, (a) is a side view of the leg-wheel type robot 100, and (b) is a plan view of the leg-wheel type robot 100.
When each drive wheel 20 is turned by turning each leg 12 around the yaw axis without grounding on the yaw axis, as shown in FIG. 9, the projection point on the floor surface of the center of gravity of the base 10 is each drive. It is separated from the center of the quadrangle formed by the ring 20 (for example, the point where the diagonal lines intersect), becomes unstable, and the possibility of falling is generated. In addition, the length of the grounding point and the rotary joint 14 which should be 0 in the grounding on the yaw axis is affected, and the driving torque of the joint motor 40 of the rotary joint 14 is increased.

  On the other hand, as shown in FIG. 10, when each leg 12 is rotated around the yaw axis while the drive wheels 20 are grounded on the yaw axis as in the present embodiment, as shown in FIG. The point of projection of the center of gravity on the floor surface is in the vicinity of the center position of the quadrangle formed by each drive wheel 20, and it is stable and the fear of falling is reduced. Further, since the length of the ground point and the rotary joint 14 is zero, the driving torque of the joint motor 40 of the rotary joint 14 can be minimized.

  On the other hand, if there are stairs on the moving path of the leg-wheel type robot 100, the horizontal plane laser light emitted from the horizontal laser 26 and the vertical surface laser lights emitted from the vertical lasers 28 and 30 are reflected by the stairs, respectively. By 32, an image including the reflected light is taken. Next, through steps S200 and S202, an image captured by the camera 32 is captured, and feature points of stairs are extracted from the captured image. Through steps S204 to S210, the width of the staircase and the actual coordinates of the nose portion are calculated based on the extracted feature points, and the landing position of the leg tip is calculated based on the calculated width of the staircase and the actual coordinates of the nose portion. Is determined.

  Further, through steps S212 to S216, sensor signals are input from the leg tip sensors 22 and 24, respectively, and the distance to the kick plate and the positional relationship between the leg tip and the tread plate are calculated. Then, through steps S218 and S220, a motor command signal is generated based on the determined landing position and the calculated both distances, and the generated motor command signal is output to the drivers 44 and 54. As a result, the driving wheel 20 rotates and the rotary joints 14 to 18 are driven, and the leg-wheel type robot 100 gets over the stairs while keeping its posture properly. Depending on the situation, the stairs are avoided and stopped. Therefore, the adaptability to the stairs is high like the legged robot.

In this way, in the present embodiment, the drive control of the leg 12 at the time of obstacle avoidance and the steering control of the drive wheel 20 at the time of traveling can be rotated around the yaw axis provided at the base of the leg 12. The joint motor 40 of the rotary joint 14 is used in common.
This eliminates the need for providing a dedicated actuator at the tip of the leg for steering control during traveling, and increases in weight and energy consumption can be suppressed as compared with the conventional case.

Further, in the present embodiment, the joint motor 40 is controlled so that the drive wheel 20 is grounded on the yaw axis during turning.
Thereby, at the time of turning, the projection point of the center of gravity of the base body 10 on the floor surface is located near the center of the quadrangle constituted by each drive wheel 20, and since it is stable, the fear of falling is reduced. Further, since the length between the ground point and the rotary joint 14 is 0, the driving torque of the joint motor 40 of the rotary joint 14 can be reduced.

Further, in the present embodiment, the image sensor including the lasers 26 to 30 and the camera 32 and the leg tip sensors 22 and 24 are provided. Based on the image captured by the camera 32 and the distance measured by the leg tip sensors 22 and 24. The stairs are recognized, and the motors 40 and 50 are controlled based on the recognition result.
Thereby, since the raising / lowering control of the leg part 12 is performed while recognizing an unknown staircase using the image sensor and the leg tip sensors 22 and 24, higher adaptability is realized with respect to the unknown staircase than in the past. be able to. In addition, since it can operate in an environment where people are active, it is suitable for home robots, personal robots, and the like that are used for acting with people.

Further, in the present embodiment, the image sensor is provided on the front surface of the base body 10, and the leg tip sensors 22 and 24 are provided on the distal ends of the leg portions 12.
As a result, it is possible to detect an object existing on the movement path of the leg wheel type robot 100 with a wide field of view, and to accurately measure the distance between the drive wheel 20 and the staircase when moving up and down the stairs.

Furthermore, in this embodiment, the lasers 26 to 30 that irradiate the laser beam at a predetermined angle with respect to the moving direction, and an image that is provided at a predetermined distance from the lasers 26 to 30 and includes the reflected light of the laser beam are captured. The camera 32 is provided, and the staircase is recognized by the light cutting method based on the image photographed by the camera 32.
Thereby, since the characteristic effective for the raising / lowering control of the leg part 12 can be detected among the characteristics of the staircase, higher adaptability can be realized for the unknown staircase.

Furthermore, in the present embodiment, the width of the staircase is calculated based on the imaging state of the reflected light of the horizontal plane laser light emitted from the horizontal laser 26, and the two vertical surface laser lights emitted from the vertical lasers 28 and 30 are calculated. The actual coordinates of the stair nosing part of the staircase are calculated based on the reflected light imaging state.
Thereby, since the characteristic more effective for the raising / lowering control of the leg part 12 can be detected among the characteristics of the staircase, higher adaptability can be realized for the unknown staircase.

Further, in the present embodiment, the distance to the stair riser plate is calculated based on the measurement result of the front leg tip sensor 22, and the positions of the drive wheels 20 and the step board of the staircase are calculated based on the measurement result of the lower leg tip sensor 24. Calculate the relationship.
Thereby, since the characteristic more effective for the raising / lowering control of the leg part 12 can be detected among the characteristics of the staircase, higher adaptability can be realized for the unknown staircase.

  In the first embodiment, the joint motor 40 of the rotary joint 14 corresponds to the actuator of the invention 1 or the first actuator of the invention 2 to 5, and the joint motor 40 of the rotary joints 16 and 18 is the invention 2 to Corresponding to the second actuator No. 5, steps S202, S220 and S222 correspond to the control means of the inventions 1 to 5.

[Second Embodiment]
Next, a second embodiment of the present invention will be described with reference to the drawings. 12 to 15 are views showing a second embodiment of the leg-wheel type robot according to the present invention.

In addition to the functions of the leg wheel type robot 100 of the first embodiment, the leg wheel type robot 100 of the present embodiment has the leg wheel type robot 100 in a state where the orientation of the base body 10 is held in a fixed direction. It has a function of running and moving in an arbitrary traveling direction, and a function of turning a leg wheel type robot without moving back and forth at a predetermined turning center position (turning in a super-confidence).
That is, the leg wheel type robot 100 according to the present embodiment is provided with an installation structure of the force sensor 82 only by adding an actuator control process (control process by executing a control program of the CPU 60) for realizing the above functions. Other configurations are the same as those of the leg-wheel type robot 100 of the first embodiment. Therefore, description of functions similar to those of the first embodiment will be omitted as appropriate, and added functional parts will be described in detail.

Hereinafter, based on FIG.12 and FIG.13, the control processing at the time of the driving | running | working performed with CPU60 of this Embodiment is demonstrated.
In each of the following travel control processes, the joints are set so that the posture of the leg-wheel type robot 100 becomes the knee flexion posture as shown in the lower diagram of FIG. 10A in the first embodiment. The motor 40 is controlled. When the posture is changed to the knee flexion posture, each joint motor 40 is controlled so that each drive wheel 20 is grounded on the yaw axis, as in the first embodiment. However, if the legs 12 interfere with each other (contact or the like) during running control in the knee flexion posture, as shown in the upper diagram of FIG. Control.

First, a description will be given of a travel control process (hereinafter, referred to as a non-directed travel control process) when the leg-wheel type robot 100 travels in a target traveling direction while maintaining the orientation of the base body 10.
The CPU 60 activates a control program stored in a predetermined area such as a ROM, and executes a non-turning traveling control process according to the control program.

  Here, the non-turning traveling control process moves the leg-wheel type robot 100 in the target traveling direction while keeping the direction of the base body 10 in a fixed direction. It is desirable to provide the camera 32, the obstacle sensors 34, 36, and the like provided on the rear side and the side. When the leg-wheel type robot 100 is moved in a direction different from the direction of the base body 10 by the camera and the obstacle sensor that can cover the rear side and the side, the environment (the terrain state, etc.) in the traveling direction is grasped. Appropriate control can be performed.

Further, the non-turning traveling control process is executed when there is a non-turning traveling control command, and the robot traveling direction (angle α) and the robot traveling direction speed Vc are input as the input. The angle of the rotary joint 14 (joint0) of each leg 12 for causing the leg wheel type robot 100 to travel in the input traveling direction with the orientation of the base body 10 held (fixed) in a fixed orientation ( Steering angle) θ _i0 (i = 0,1,2,3,...), Rotational angular velocity ω _i (i = 0,1,2,3,...) Of each drive wheel 20 of each leg 12 Is calculated and a command is given to each actuator.

As described in the first embodiment, the leg-wheel type robot 100 according to the present embodiment has a total of four leg portions 12 in a pair of left and right in the front of the base 10 and a pair of left and right in the rear. .
Therefore, here, the steering angle θ _i0 of the driving wheel 20 of each leg 12 is viewed from the upper surface side of the base 10, the left front wheel steering angle θ ₀₀ , the right front wheel steering angle θ ₁₀ , and the left rear wheel steering angle θ _20. The right rear wheel steering angle θ ₃₀ is set. Incidentally, the rotation joint 14 when rotated each leg 12 about the yaw axis, as viewed from the top side of the base body 10, a left front-wheel steering angle theta ₀₀ and the right rear wheel steer angle theta ₃₀ is counterclockwise The direction is a positive direction, and the right front wheel steering angle θ ₁₀ and the left rear wheel steering angle θ ₂₀ are clockwise directions.

Further, the rotational angular velocities ω _i of the drive wheels 20 of the respective leg portions 12 are defined as a left front wheel rotational angular velocity ω ₀ , a right front wheel rotational angular velocity ω ₁ , a left rear wheel rotational angular velocity ω ₂ , and a right rear wheel rotational angular velocity ω ₃ .
Further, the linear velocity V _i (i = 0, 1, 2, 3,...) Of the driving wheel 20 of each leg 12 is set to the left front wheel linear velocity V ₀ , the right front wheel linear velocity V ₁ , and the left rear wheel lane. A speed V ₂ and a right rear wheel linear speed V ₃ are used.
Here, FIGS. 12A to 12C are diagrams illustrating an example of the traveling state of the leg-wheel type robot 100 during the non-turning traveling control. 12A to 12C are views of the leg-wheel type robot 100 as viewed from the upper surface side, and the black semicircle mark attached to each drive wheel 20 is “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = 0 [°] ”indicates a reference direction.

Further, the rotation direction of the left and right front wheels traveling in the direction with the mark is the positive rotation direction, and the rotation direction of the left and right rear wheels traveling in the direction without the mark is the positive rotation direction.
First, based on FIG. 12A, the leg-wheel type robot 100 is made to travel straight in the direction (forward direction) in which the base body 10 is facing in a state where the direction of the base body 10 is held (fixed) in a certain direction. The non-turning traveling control process will be described.

In the present embodiment, when the direction in which the base body 10 is facing (forward direction) is the traveling direction of the robot, the angle α representing the traveling direction is set to “0 [°]”. Then, α corresponding to each traveling direction is determined with reference to 0 [°] in the forward direction.
Here, since the leg wheel type robot 100 is caused to advance straight forward of the base body 10, “0 [°]” is input as the traveling direction α, and the traveling direction speed Vc is further input.

When the traveling direction α (0 [°]) and the traveling direction speed Vc are input, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ of each driving wheel 20 and the linear speed V ₀ of each driving wheel 20 V ₁ , V ₂ and V ₃ are calculated.
In order to cause the leg wheel type robot 100 to travel straight forward while maintaining the orientation of the base body 10, as shown by the arrows extending from the drive wheels 20 in FIG. It is necessary to align all the traveling directions in the direction in which the base body 10 is facing (forward direction). Accordingly, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated as, for example, “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = 0 [°]”.

Further, in this case, the linear velocities V ₀ , V ₁ , V ₂ , and V ₃ of the drive wheels 20 for running straight ahead are “V ₀ = V ₁ = Vc” and “V ₂ = V ₃ = −Vc”. Is calculated.
If the legs 12 do not interfere with each other, for example, “θ ₀₀ = θ ₁₀ = 0 [°]”, “θ ₂₀ = θ ₃₀ = π (180 [°])” or −π (−180 [° ]) ”,“ V ₀ = V ₁ = V ₂ = V ₃ = Vc ”, etc.

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the following equation (8).

_{ω i = 2V i / D ...} (8)

However, in the above formula (8), D is a wheel diameter.

Once the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} drive wheels 20 are calculated, 20 current steering angles and rotational angular velocities are obtained. Then, a command value for driving the joint motor 40 of the rotary joint 14 is calculated from the current steering angle and the calculated steering angle for moving straight forward. Further, a command value for driving the wheel motor 50 of the drive wheel 20 is calculated from the current rotation angular velocity and the rotation angular velocity when the vehicle travels straight in the forward direction.

  When the steering control command value and the speed control command value of each drive wheel 20 are calculated in this way, these command values are input to the drivers of the motors. Then, when each joint motor 40 and wheel motor 50 are driven based on the input command value, the steering angle and the rotational angular velocity of each drive wheel 20 of the leg wheel type robot 100 are thereby changed, and the leg wheel type robot is changed. In a state where the orientation of the base body 10 is maintained, the vehicle 100 travels straight in the facing direction (forward direction).

When the leg-wheel type robot 100 is allowed to travel straight in the direction opposite to the direction in which the base body 10 is facing (directly rearward) while maintaining the orientation of the base body 10, the driving wheel 20 and the driving wheel 20 are used. The rotation direction may be opposite.
For example, “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = 0 [°]” and “V ₀ = V ₁ = V ₂ = V ₃ = −Vc” or “θ ₀₀ = θ ₁₀ = 0 [°] “Θ ₂₀ = θ ₃₀ = π or −π [°]”, “V ₀ = V ₁ = −Vc”, “V ₂ = V ₃ = Vc”, and the like.

Next, based on FIG. 12 (b), in the state where the direction of the base body 10 is maintained, the leg-wheel type robot 100 travels straight ahead in the diagonally forward direction with respect to the direction in which the base wheel 10 is directed. The control process will be described.
Here, since the leg-wheel type robot 100 moves straight forward in the diagonally right direction of the base body 10, the traveling direction α is “α (−90 <α <0) [°] (where counterclockwise is the positive direction)”. Is input, and the traveling direction speed Vc is further input.

When the traveling direction α [°] and the traveling direction speed Vc are input, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ of each driving wheel 20 and the linear speeds V ₀ , V of each driving wheel 20 are input. ₁ , V ₂ and V ₃ are calculated.
In order to cause the leg wheel type robot 100 to travel straight forward in the diagonally forward right direction (α [°]) while maintaining the orientation of the base body 10, the arrows extending from the drive wheels 20 in FIG. As shown, it is necessary to align all the traveling directions of the drive wheels 20 in the diagonally forward right direction with respect to the direction in which the base body 10 is directed.

Therefore, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ are calculated as, for example, “θ ₀₀ = θ ₃₀ = α [°]”, “θ ₁₀ = θ ₂₀ = −α [°]”.
Further, in this case, the linear velocities V ₀ , V ₁ , V ₂ , and V ₃ of the drive wheels 20 for running straight ahead are “V ₀ = V ₁ = Vc” and “V ₂ = V ₃ = −Vc”. Is calculated.
If the legs 12 do not interfere with each other, for example, “θ ₀₀ = θ ₃₀ = α [°]”, “θ ₁₀ = θ ₂₀ = −α−π [°]”, “V ₀ = V ₂ = Vc ", may be as a combination, such as" V ₁ = V ₃ = -Vc ".

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
When the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} respective drive wheels 20 are calculated, next, the straight traveling in the forward direction is performed. Similarly, the command value for driving the joint motor 40 of the rotary joint 14 and the command value for driving the wheel motor 50 of the drive wheel 20 are calculated.

  When the steering control command value and the speed control command value of each drive wheel 20 are calculated in this way, these command values are input to the drivers of the motors. Then, when each joint motor 40 and wheel motor 50 are driven based on the input command value, the steering angle and the rotational angular velocity of each drive wheel 20 of the leg wheel type robot 100 are thereby changed, and the leg wheel type robot is changed. In a state where the orientation of the base body 10 is maintained, the vehicle 100 travels straight forward in a diagonally forward right direction with respect to the facing direction.

  When the leg-wheel type robot 100 travels straight diagonally to the left and backward with respect to the direction in which the base body 10 is held, each drive wheel 20 in the right diagonal forward direction is used. What is necessary is just to make the rotation direction opposite. Further, when moving straight ahead in the diagonally left direction, the traveling direction α may be set in a range of “0 <α <90 [°]”, and the sign of the steering angle may be opposite. Further, in the case where the vehicle travels straight diagonally to the right, the rotational direction of each drive wheel 20 in the diagonally forward left direction may be opposite to that of the front.

Next, based on FIG. 12 (c), in the state in which the base body 10 is held, the non-turning traveling control in the case where the leg-wheel type robot 100 travels straight in the left lateral direction with respect to the facing direction. Processing will be described.
Here, since the leg-wheel type robot 100 is moved straight in the left lateral direction of the base body 10, “α (90 (π / 2)) [°]” is input as the traveling direction α, and the traveling direction velocity Vc is Entered.

When the traveling direction α [°] and the traveling direction speed Vc are input, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ of each driving wheel 20 and the linear speeds V ₀ , V of each driving wheel 20 are input. ₁ , V ₂ and V ₃ are calculated.
In order to cause the leg wheel type robot 100 to travel straight in the left lateral direction (the direction of π / 2 [°]) with respect to the direction in which the base body 10 is held, FIG. ), It is necessary to align all the traveling directions of the driving wheels 20 in the left lateral direction.

Accordingly, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ are calculated as, for example, “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = π / 2 [°]”.
In this case, the linear velocities V ₀ , V ₁ , V ₂ , and V ₃ of the drive wheels 20 for running straight ahead are “V ₀ = V ₂ = Vc” and “V ₁ = V ₃ = −Vc”. Is calculated.
If the legs 12 do not interfere with each other, for example, “θ ₀₀ = θ ₂₀ = π / 2 [°]”, “θ ₁₀ = θ ₃₀ = −π / 2 [°]”, “V ₀ = A combination such as “V ₁ = V ₂ = V ₃ = Vc” may be adopted.

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
When the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} respective drive wheels 20 are calculated, next, the straight traveling in the forward direction is performed. Similarly, the command value for driving the joint motor 40 of the rotary joint 14 and the command value for driving the wheel motor 50 of the drive wheel 20 are calculated.

  When the steering control command value and the speed control command value of each drive wheel 20 are calculated in this way, these command values are input to the drivers of the motors. Then, when each joint motor 40 and wheel motor 50 are driven based on the input command value, the steering angle and the rotational angular velocity of each drive wheel 20 of the leg wheel type robot 100 are thereby changed, and the leg wheel type robot is changed. In a state in which the orientation of the base body 10 is maintained, the vehicle 100 travels straight in the left lateral direction with respect to the facing direction.

When the leg-wheel type robot 100 travels straight in the right lateral direction with respect to the direction in which the base body 10 is held, the rotation of each driving wheel 20 in the left lateral direction is performed. The direction may be opposite.
Next, a description will be given of a travel control process (hereinafter referred to as a “super-revolution turn control process”) in which the leg-wheel robot 100 turns without moving back and forth at a predetermined turn center position (hereinafter referred to as “super-revolution turn control process”).

The CPU 60 activates a control program stored in a predetermined area such as a ROM, and executes super turning control processing according to the control program.
Here, the super-revolution turning control process causes the leg wheel type robot 100 to perform a turn operation equivalent to super-revolution performed by a vehicle such as a power shovel or a tank having a crawler mechanism.
In addition, super turning is a turning method in which a vehicle having a crawler mechanism rotates left and right crawlers in the opposite directions at the same speed so that the vehicle body does not move forward and backward and the direction of the vehicle body changes. This is a turning method that can be realized not only in the crawler mechanism but also in a vehicle having at least two independent drive wheels on the left and right.

In the present embodiment, specifically, it is executed when there is a super turning control command, and the turning angular velocity Ω of the robot and the turning center (xc, yc) are used as inputs to turn the leg wheel type robot 100 into the turning center. Angles (steering angles) θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ of the rotary joints 14 (joint 0) of the respective leg portions 12 for driving each of the leg portions 12 in order to make a super turn at (xc, yc) The rotational angular velocities ω ₀ , ω ₁ , ω ₂ , and ω ₃ of the wheel 20 are calculated, and commands are given to each actuator.

  In order to cause the leg wheel type robot 100 to make a super turn, when the base body 10 is rotated about the yaw axis at the turning center (xc, yc), the rotation about the yaw axis at the time of steering of each drive wheel 20 is performed. The joint motor 40 of each rotary joint 14 is controlled so that the direction of movement at the contact point between the arc trajectory drawn by the center and the rotation center of each drive wheel 20 matches the traveling direction of each drive wheel 20, and each drive It is necessary to control each wheel motor 50 so that the wheel 20 rotates at a constant speed in the rotation direction corresponding to the movement direction.

Here, FIGS. 13A and 13B show the leg-wheel type robot 100 at the time of super-superior turning control with the coordinates (0, 0) and coordinates (xc, yc) of the base body 10 as the rotation center. It is a figure which shows a driving | running | working state.
In FIGS. 13A and 13B, in the plane when the base body 10 is viewed from the upper surface side, the longitudinal axis is the x-axis and the axis perpendicular thereto is the y-axis. The coordinates are (x, y) = (0, 0).

First, based on FIG. 13 (a), a description will be given of the super-symbol turning control process when the coordinates (0, 0) of the center position of the base body 10 is set as the turning center.
In this case, the turning angular velocity Ω and the turning center coordinates (0, 0) are input.
When the turning angular velocity Ω and the turning center coordinates (0, 0) are input, the angles φ ₀ , φ ₁ , φ formed by the x axis of the base 10 and the rotation center of each drive wheel 20 based on the following equation (9). _2, phi ₃ is calculated.

tan φ = W _t / W _b (9)

However, the above equation (9) is an equation in the case where the center coordinate (0, 0) of the base body 10 is set as the turning center. In the above equation (9), W _t is a tread (wheel spacing), and W _b is a wheel base.

Note that φ _{0 to} φ ₃ are the minimum angles among the angles formed between the x-axis and the rotation center of each drive wheel 20, and the positive direction of φ differs from the positive direction of Ω depending on the drive wheels.
Since W _t and W _b are known (information is stored in advance), the angles φ ₀ , φ ₁ , φ ₂ , and φ ₃ can be calculated from the above equation (9).
Since the turning center is the center coordinate (0, 0) of the base body 10, the angles φ ₀ , φ ₁ , φ ₂ , and φ ₃ are all equal angles “φ ₀ = φ ₁ = φ ₂ = φ ₃ = φ ”.

In addition, when the base body 10 rotates around the yaw axis at the turning coordinates (0, 0), the movement direction at the contact point between the circular arc drawn by the rotation center around the yaw axis and the rotation center at the time of steering of each drive wheel 20 is: As indicated by the arrow lines extending from each drive wheel 20 in FIG. 13A, the direction is a tangential direction (in the direction of the arrow line in the figure) passing through each rotation center on the circular arc track.
Once the angle “φ ₀ = φ ₁ = φ ₂ = φ ₃ = φ” with respect to the rotation center of each drive wheel 20 is calculated, the movement direction of each rotation center and the traveling direction of each drive wheel 20 are matched. Steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated.

In order to make the movement direction of each rotation center coincide with the traveling direction of each drive wheel 20, as shown in FIG. 13 (a), a line connecting the traveling direction of each drive wheel 20, the turning center and the rotation center. min angle between the perpendicular (π / 2 (90 [° ])) and so as well if the steering, therefore, the steering angle _{θ 00, θ 10, θ 20} , θ 30 is "θ ₀₀ = θ ₁₀ = Θ ₂₀ = θ ₃₀ = − (π / 2−φ) ”.
On the other hand, the distances L ₀ , L ₁ , L ₂ , L ₃ between the rotation center around the yaw axis and the turning center (0, 0) during steering of each drive wheel 20 are calculated based on the following equation (10). .

However, the above formula (10) is a formula in the case where the center coordinate (0, 0) of the base body 10 is set as the turning center.
From the tread W _t and the wheel base W _b , the coordinates of the rotation center of the left front wheel are (W _b / 2, W _t / 2), and the coordinates of the rotation center of the right front wheel are (W _b / 2, −W _t / 2) The coordinates of the center of rotation of the left rear wheel are represented as (−W _b / 2, W _t / 2), and the coordinates of the center of rotation of the right rear wheel are represented as (−W _b / 2, −W _t / 2). Can do.

Since the turning center is the center coordinate (0, 0) of the base body 10, the distances L ₀ , L ₁ , L ₂ , L ₃ are equal distances “L ₀ = L ₁ = L ₂ = L ₃ = L”. It becomes.
Once the distances L ₀ , L ₁ , L ₂ , and L ₃ are calculated, the linear velocities V ₀ , V ₀ , and V of each drive wheel 20 are then calculated from these distances L and the turning angular velocity Ω based on the following equation (11). V ₁ , V ₂ and V ₃ are calculated.

V ₀ = V ₁ = V ₂ = V ₃ = LΩ (11)

However, the above expression (11) is an expression when the center coordinate (0, 0) of the base body 10 is set as the turning center.

Here, as described above, the rotation direction of the left and right front wheels traveling in the direction with the black semicircle mark is the positive rotation direction, and the rotation direction of the left and right rear wheels traveling in the direction without the mark is the positive rotation direction.
Further, since the directions of the drive wheels 20 are as shown in FIG. 13A, the linear velocities V ₀ , V ₁ , V ₂ , and V ₃ are “V ₀ = V ₃ = −LΩ”, “V ₁ = V ₂ = LΩ ”is calculated.
Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).

Once the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ of each drive wheel 20 are calculated, the current steering of each drive wheel 20 is then performed. Get the angle and rotational angular velocity. Then, a command value for driving the joint motor 40 of the rotary joint 14 is calculated from the current steering angle and the calculated steering angle. Further, a command value for driving the wheel motor 50 of the drive wheel 20 is calculated from the current rotation angular velocity and the calculated rotation angular velocity.

  When the steering control command value and the speed control command value of each drive wheel 20 are calculated in this way, these command values are input to the drivers of the motors. Then, when each joint motor 40 and wheel motor 50 are driven based on the input command value, the steering angle and the rotational angular velocity of each drive wheel 20 of the leg wheel type robot 100 are thereby changed, and the center of the base body 10 is changed. With the coordinates (0, 0) as the turning center, the leg-wheel type robot 100 turns on the spot without making a back-and-forth movement.

Next, the super turning control process when coordinates other than the center coordinate (0, 0) of the base 10 are set as the turning center will be described. Hereinafter, this super turning control process is referred to as a turning center offset type super turning control process.
In this case, the turning angular velocity Ω and turning center coordinates (xc, yc) ≠ (0, 0) which are coordinates other than the center coordinates (0, 0) of the base body 10 are input.

When the turning angular velocity Ω and the turning center coordinates (xc, yc) are input, the angles φ ₀ , φ ₁ formed by the x axis of the base 10 and the rotation center of each drive wheel 20 based on the following equation (12). , Φ ₂ , φ ₃ are calculated.

tan φ ₀ = (W _t / 2-yc) / (W _b / 2-xc)
tan φ ₁ = (W _t / 2 + yc) / (W _b / 2-xc)
tanφ ₂ = (W _t / 2-yc) / (W _b / 2 + xc)
tan φ ₃ = (W _t / 2 + yc) / (W _b / 2 + xc) (12)

However, the above formula (12) is a formula when the leg portions 12 of the leg-wheel type robot 100 are a total of four pairs of left and right pairs in front of the base 10 and left and right pairs in the rear.

Specifically, the input turning center coordinates (xc, yc) and known W _t and W _b are substituted into the above equation (12), and the x axis of the base 10 and the rotation center of each driving wheel 20 are substituted. Are calculated. Φ ₀ , φ ₁ , φ ₂ , φ ₃ are calculated.
Since the turning center is a coordinate other than the center coordinate of the base body 10, the angles φ ₀ , φ ₁ , φ ₂ , and φ ₃ are different from each other.

  Further, when the base body 10 rotates around the yaw axis at the turning center coordinates (xc, yc), the direction of motion at the contact point between the circular arc drawn by the rotation center around the yaw axis and the rotation center at the time of steering of each drive wheel 20. As shown by the arrow lines extending from the respective drive wheels 20 in FIG. 13 (b), they are tangential directions (in the direction of the arrow lines in the figure) passing through the respective rotation centers on the respective circular arc tracks. Further, since the distance between the turning center and the rotation center of each drive wheel 20 is different, the arc trajectory drawn by each rotation center is also different.

Once the angles φ ₀ , φ ₁ , φ ₂ , and φ ₃ with respect to the rotation center of each drive wheel 20 are calculated, next, steering for making the movement direction of each rotation wheel coincide with the traveling direction of each drive wheel 20. The angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated.
In order to make the movement direction of each rotation center coincide with the traveling direction of each drive wheel 20, as shown in FIG. 13B, a line connecting the rotation direction of each drive wheel 20, the turning center and the rotation center. It is only necessary to steer so that the angle formed by the minute becomes a right angle (π / 2 (90 [°])). Therefore, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are “θ _i0 = − ( π / 2-φ _i) is calculated as (i = 0, 1, 2, 3) ".

On the other hand, the distances L ₀ , L ₁ , L ₂ , L ₃ between the rotation center around the yaw axis and the turning center (xc, yc) at the time of steering of each drive wheel 20 are calculated based on the following equation (13). .

As described above, since the turning center is a coordinate other than the center coordinates (0, 0) of the base body 10, the distances L ₀ , L ₁ , L ₂ , and L ₃ are different distances.
Once the distances L ₀ , L ₁ , L ₂ , and L ₃ are calculated, the linear velocities V ₀ , V ₀ of the drive wheels 20 are then calculated from these distances and the turning angular velocity Ω based on the following equation (14). ₁ , V ₂ and V ₃ are calculated.

_{| V i | = | L i} Ω | ... (14)

Therefore, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are “V ₀ = −L ₀ Ω”, “V ₁ = L ₁ Ω”, “V ₂ = L ₂ Ω”, “V ₃ = − L ₃ Ω ”.

Furthermore, the linear velocity _{V 0, V 1, V 2} , V 3 in accordance with the above equation (8), the rotation angular velocity _{ω 0, ω 1, ω 2} , is converted to omega _3.
Once the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ of each drive wheel 20 are calculated, the current steering of each drive wheel 20 is then performed. Get the angle and rotational angular velocity. Then, a command value for driving the joint motor 40 of the rotary joint 14 is calculated from the current steering angle and the calculated steering angle. Further, a command value for driving the wheel motor 50 of the drive wheel 20 is calculated from the current rotation angular velocity and the calculated rotation angular velocity.

  When the steering control command value and the speed control command value of each drive wheel 20 are calculated in this way, these command values are input to the drivers of the motors. Then, when each joint motor 40 and wheel motor 50 are driven based on the input command value, the steering angle and the rotational angular velocity of each drive wheel 20 of the leg wheel type robot 100 are thereby changed. With the coordinates (xc, yc) other than the position (0, 0) as the turning center, the leg-wheel robot 100 turns (super turning) without moving back and forth.

Next, the operation of the present embodiment will be described with reference to FIGS.
Here, FIGS. 14A and 14B are diagrams illustrating an example of a travel route of the leg-wheel type robot 100. FIG. FIG. 15 is a diagram illustrating an example of the center of gravity position of the robot.
If the leg-wheel type robot 100 determines that there is no obstacle on its own movement path based on the sensor signals from the obstacle sensors 34, 36, etc. Switching from the leg movement mode to be used to the wheel traveling movement mode using the drive wheels 20 is performed.

On a flat ground, the leg-wheel type robot 100 can move in the above-described non-turned traveling in addition to the wheel traveling described in the first embodiment. Further, in addition to the turning described in the first embodiment, the above-described super-spinning and the turning center offset type super-spinning can be performed.
Further, when the wheels are running and turning, the joint motor 40 is controlled so that the leg-wheel type robot 100 assumes a knee bending posture, and the joint motor 40 is controlled so that each drive wheel 20 is grounded on the yaw axis. The

First, the operation of the leg wheel type robot 100 during the non-turning traveling control will be described.
Here, it is assumed that the leg wheel type robot 100 travels and moves along a passage as shown in FIG. 14A and 14B are overhead views of a part of the travel route (passage) as seen from directly above.
As shown in FIG. 14 (a), the width of the passage is initially narrow enough to pass through in the direction (front direction) in which the base body 10 is facing, so first, before entering the passage, The direction in which the passage extends and the direction of the base 10 are matched, and the entry position and the entry angle are adjusted.

Then, the non-turning traveling control command is input, and the leg wheel type robot 100 is shifted to the non-turning traveling control mode. As a result, the leg wheel type robot 100 controls the CPU 60 to travel and move in the target traveling direction while maintaining the orientation of the base 10.
First, since the leg wheel type robot 100 is desired to move straight forward, the robot traveling direction α = 0 [°] and the robot traveling direction speed Vc are input. Thereby, the steering angle of each drive wheel 20 is calculated as “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = 0 [°]”, and the linear velocity of each drive wheel 20 is “V ₀ = V ₁ = V _2”. = V ₃ = Vc ”.

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
When the CPU 60 calculates the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ , and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ of the drive wheels 20, the angle capture I / F 62 is then obtained. The current steering angle and rotational angular velocity of each drive wheel 20 are acquired. Here, a difference value between the acquired steering angle and the calculated steering angle is calculated, and a command value for driving the joint motor 40 of the rotary joint 14 is calculated based on the difference value. Further, a difference value between the acquired current rotation angular velocity and the calculated rotation angular velocity is calculated, and a command value for driving the wheel motor 50 of the drive wheel 20 is calculated based on the difference value.

  The CPU 60 inputs the calculated steering control command value and speed control command value of each drive wheel 20 to the driver of each motor. With this command value, each joint motor 40 is driven, and the rotary joint 14 of each leg 12 rotates about the yaw axis to change to a target steering angle. Thereafter, each wheel motor 50 is driven, and each drive wheel 20 is rotationally driven at a rotational angular velocity corresponding to the command value. As a result, the leg wheel type robot 100 enters the passage in the direction (front direction) in which the base body 10 is facing while keeping the orientation of the base body 10 and travels straight in the passage.

When the leg-wheel type robot 100 goes straight forward for a while, the passage bends to the right at a substantially right angle and further widens the road width. Control to move straight in the direction. That is, α = −90 (−π / 2) [°] is input as the traveling direction, and further the traveling direction velocity Vc is input.
Thereby, “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = α [°]” is calculated as the steering angle, and “V ₀ = V ₂ = −Vc”, “V ₁ = V ₃ ” as the linear velocities. = Vc "is calculated.

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
The CPU 60 acquires the current steering angle and rotational angular velocity of each drive wheel 20, and calculates each command value from these and the calculated steering angle and rotational angular velocity. By this command value, each joint motor 40 and each wheel motor 50 are driven, so that the leg wheel type robot 100 holds the direction of the base body 10 in the right lateral direction with respect to the direction in which it is directed. Drive straight ahead. Thus, the leg wheel type robot 100 can be caused to travel to the bent point without turning at the corner of the passage.

When the leg wheel type robot 100 goes straight to the right side for a while, the passage bends diagonally downward to the right. Control to move straight. That is, α (passage angle <0) [°] is input as the traveling direction, and further the traveling direction velocity Vc is input.
Thereby, for example, “θ ₀₀ = θ ₃₀ = π + α [°]” and “θ ₁₀ = θ ₂₀ = − (π + α) [°]” are calculated as the steering angle, and “V ₀ = V “ ₁ = −Vc” and “V ₂ = V ₃ = Vc” are calculated.

Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
Then, the current steering angle and rotation angular velocity of each drive wheel 20 are acquired, and each command value is calculated from these and the calculated steering angle and rotation angular velocity. By this command value, each joint motor 40 and each wheel motor 50 are driven, so that the leg wheel type robot 100 holds the orientation of the base body 10 in the diagonally rightward direction with respect to the direction in which the base 10 is facing. Go straight ahead.

Next, the operation of the leg wheel type robot 100 at the time of super-trust turning control will be described.
Here, it is assumed that the leg wheel type robot 100 travels and moves along a passage as shown in FIG.
As shown in FIG. 14 (b), the passage extends straight at first, then turns right at a substantially right angle, and reaches a dead end.

First, the direction in which the passage extends and the direction of the base body 10 are matched and the approach position is finely adjusted, and then the leg-wheel type robot 100 is caused to travel straight ahead in the forward direction. Thereby, the leg-wheel type robot 100 enters the passage and travels straight in the passage.
When the leg wheel type robot 100 goes straight for a while, the leg wheel type robot 100 eventually reaches the turning corner, and therefore turns clockwise to change the direction of the base body 10 so that the leg wheel type robot 100 can advance.

  In the case of the corner shown in FIG. 14B, it is possible to make a right turn even with a turning motion accompanied by a movement. For this reason, a super-trust turning control command is input, and the leg wheel type robot 100 is shifted to the super-trust turning control mode by this command. As a result, the leg wheel type robot 100 performs a control process in which the CPU 60 turns the robot at a predetermined turning center position without moving back and forth.

First, a turning angular velocity Ω (an angular velocity for rotating the base body 10 clockwise) and a turning center coordinate (0, 0) are input. Further, since the right turn is performed at a substantially right angle, a turning angle of −90 [°] is input.
When the turning angular velocity Ω, the turning center coordinates (0, 0), and the turning angle (−90 [°]) are input, the x axis of the base 10 and the rotation center of each driving wheel 20 are determined based on the above equation (9). Are calculated. Φ ₀ , φ ₁ , φ ₂ , φ ₃ are calculated.

Once the angle “φ ₀ = φ ₁ = φ ₂ = φ ₃ = φ” with respect to the rotation center of each drive wheel 20 is calculated, the movement direction of each rotation center and the traveling direction of each drive wheel 20 are matched. Steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated.
Specifically, “θ ₀₀ = θ ₁₀ = θ ₂₀ = θ ₃₀ = − (π / 2−φ)” is calculated as the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ .

Further, based on the above equation (10), distances L ₀ , L ₁ , L ₂ , L ₃ between the rotation center around the yaw axis and the turning center (0, 0) at the time of steering of each drive wheel 20 are calculated. .
Since the turning center is the center coordinates (0, 0) of the base body 10, the distances L ₀ , L ₁ , L ₂ , L ₃ are equal distances “L ₀ = L ₁ = L ₂ = L ₃ = L”. It becomes.
Next, the linear velocities V ₀ , V ₁ , V ₂ , and V ₃ of the drive wheels 20 are calculated from the distance L and the turning angular velocity Ω based on the above equation (11).

Here, since the leg-wheel type robot 100 is turned clockwise (Ω <0), the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are “V ₁ = V ₂ = LΩ”, “V ₀ = V ₃ = −LΩ ”.
Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).
Once the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} drive wheels 20 are calculated, then, via the angle capture I / F 62. Thus, the current steering angle and rotational angular velocity of each drive wheel 20 are acquired. Here, a difference value between the acquired steering angle and the calculated steering angle is calculated, and a command value for driving the joint motor 40 of the rotary joint 14 is calculated based on the difference value. Further, a difference value between the acquired current rotation angular velocity and the calculated rotation angular velocity is calculated, and a command value for driving the wheel motor 50 of the drive wheel 20 is calculated based on the difference value.

  The CPU 60 inputs the calculated steering control command value and speed control command value of each drive wheel 20 to the driver of each motor. With this command value, each joint motor 40 is driven, and the rotary joint 14 of each leg 12 rotates about the yaw axis to change to a target steering angle. Thereafter, each wheel motor 50 is driven, and each drive wheel 20 is rotationally driven at a rotational angular velocity corresponding to the command value. As a result, the leg-wheel type robot 100 turns 90 [°] clockwise (superficial turning) on the spot without moving back and forth with the center coordinate (0, 0) of the base body 10 as the turning center.

When the leg wheel type robot 100 turns 90 [°] clockwise and turns right, the robot travels forward in the forward direction along the passage.
As shown in FIG. 14B, the right-turned passage is a dead end, so the leg-wheel robot 100 eventually reaches the dead end of the passage.
Since the leg-wheel type robot 100 cannot proceed any further (ascertained by various sensors), the leg-wheel type robot 100 turns 180 [°] and turns back the passage.

Here, since the width of the passage is not wide enough to make a turning motion (for example, U-turn) with back-and-forth movement, as in the case of the right turn, the mode shifts to the super turning control mode, It turns 180 [°], changes the direction of the leg wheel type robot 100, and turns it back.
Further, as shown in FIG. 14B, since the passage width is the minimum width of rotation, it is necessary to turn the leg wheel type robot 100 with the minimum turning radius. Therefore, the turning center coordinates (0, 0), the turning angular velocity Ω (the angular velocity for rotating the base 10 clockwise), and the turning angle 180 [°] are input. Note that the posture of each leg 12 of the leg wheel type robot 100 is changed to the knee extension posture so that the leg 12 does not hit the passage.

When the turning center coordinates (0, 0), turning angular velocity Ω, and turning angle 180 [°] are input, the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ _{30 are set} as “θ _{00 = θ 10 = θ 20 =} θ 30 = - (π / 2-φ) "is calculated as the linear velocity _{V 0, V 1, V 2} , V 3, " V ₁ = V ₂ = LΩ "," V ₀ = V ₃ = −LΩ ”is calculated.
Further, the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are converted into rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω ₃ according to the above equation (8).

When the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} drive wheels 20 are calculated, the command values are calculated, and the joint motor 40 and The wheel motor 50 is driven.
As a result, the leg wheel type robot 100 turns 180 [°] clockwise (super turning) on the spot without moving back and forth, with the center coordinate (0, 0) of the base body 10 as the turning center, and in the back direction. Change direction.

  Here, for example, when the leg-wheel type robot 100 is travel-controlled manually by remote operation, or when it is automatically controlled and has various sensors and cameras on the back side of the base body 10, no change is made. By running control, it is also possible to run straight back while turning forward and turn the path back. However, in the case of automatic control and when various sensors are provided only on the front side of the base body 10, it is necessary to match the direction of the base body 10 with the traveling direction. Accordingly, in the latter case, the super-symbol turning is a useful turning means.

Next, the operation of the leg-wheel type robot 100 at the time of turning center offset type super-revolution turning control will be described.
Now, it is assumed that the mode is shifted to the wheel traveling mode, each joint motor 40 of the leg wheel type robot 100 is controlled, and the posture of the robot is changed to the knee flexion posture. At this time, it is assumed that the center of gravity of the robot deviates from the coordinates of the center position of the substrate 10 as shown in FIG.

In this case, since the position of the center of gravity is deviated from the coordinates (0, 0) of the center position of the base body 10, if super turning is performed with the center position as the turning center, the turning becomes unbalanced and a problem may occur. There is.
In such a case, in the super-revolution turn control command mode of the present embodiment, since the super-revolution can be made at an arbitrary turning center, the turning angular velocity Ω is input and the leg wheel in the base body 10 is input. The coordinates (xg, yg) corresponding to the center of gravity position of the robot 100 are input as the turning center coordinates (xc, yc).

When the turning angular velocity Ω and the turning center coordinates (xc, yc) = (xg, yg) are input, the x axis of the base 10 and the rotation center of each drive wheel 20 are formed based on the above equation (12). The angles φ ₀ , φ ₁ , φ ₂ , φ ₃ are calculated.
Once the angles φ ₀ , φ ₁ , φ ₂ , and φ ₃ with respect to the rotation center of each drive wheel 20 are calculated, next, steering for making the movement direction of each rotation wheel coincide with the traveling direction of each drive wheel 20. The angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated.

The steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , and θ ₃₀ are calculated as “θ _i0 = − (π / 2−φ _i ) (i = 0, 1, 2, 3)”.
Next, distances L ₀ , L ₁ , L ₂ , and L ₃ between the rotation center around the yaw axis and the turning center (xc, yc) during the steering of each drive wheel 20 are calculated based on the above equation (13). .
Once the distances L ₀ , L ₁ , L ₂ , L ₃ are calculated, the linear velocities V ₀ , V ₀ of the drive wheels 20 are then calculated from these distances and the turning angular velocity Ω based on the above equation (14). ₁ , V ₂ and V ₃ are calculated.

Here, assuming that the leg-wheel type robot 100 is turned counterclockwise (Ω> 0), the linear velocities V ₀ , V ₁ , V ₂ , V ₃ are “V ₀ = −L ₀ Ω”, “V ₁ = L ₁ Ω ”,“ V ₂ = L ₂ Ω ”, and“ V ₃ = −L ₃ Ω ”.
Furthermore, the linear velocity _{V 0, V 1, V 2} , V 3 in accordance with the above equation (8), the rotation angular velocity _{ω 0, ω 1, ω 2} , is converted to omega _3.

Once the steering angles θ ₀₀ , θ ₁₀ , θ ₂₀ , θ ₃₀ and the rotational angular velocities ω ₀ , ω ₁ , ω ₂ , ω _{3 of the} drive wheels 20 are calculated, then, via the angle capture I / F 62. Thus, the current steering angle and rotational angular velocity of each drive wheel 20 are acquired. Here, a difference value between the acquired steering angle and the calculated steering angle is calculated, and a command value for driving the joint motor 40 of the rotary joint 14 is calculated based on the difference value. Further, a difference value between the acquired current rotation angular velocity and the calculated rotation angular velocity is calculated, and a command value for driving the wheel motor 50 of the drive wheel 20 is calculated based on the difference value.

  The CPU 60 inputs the calculated steering control command value and speed control command value of each drive wheel 20 to the driver of each motor. By this command value, each joint motor 40 and each wheel motor 50 are driven, whereby the leg wheel type robot 100 is balanced with the coordinates (xg, yg) corresponding to the center of gravity coordinates in the base body 10 as the turning center. Turn the superstrate counterclockwise in the state.

Thus, in the present embodiment, the joint motor 40 is controlled so that the driving wheel 20 is grounded on the yaw axis while the posture of the leg-wheel type robot 100 is set to the knee bending posture during traveling control.
As a result, when traveling, the projection point of the center of gravity of the base body 10 on the floor surface is located near the center of the quadrangle formed by the drive wheels 20, and since it is stable, the fear of falling is reduced. Further, since the length between the ground point and the rotary joint 14 is 0, the driving torque of the joint motor 40 of the rotary joint 14 can be reduced.

Further, in the present embodiment, the joint motor 40 and the wheel motor 50 are controlled so that the leg-wheel type robot 100 travels in the target traveling direction while maintaining the orientation of the base body 10 in a certain direction.
Thereby, since it is possible to move in any direction without turning (without changing the direction), it is possible to realize quick movement in each direction, and it is impossible to make a turn because each component of the leg-wheel type robot 100 is hindered. Even in areas where it is difficult to change the direction of the leg-wheel type robot, such as a narrow and intricate area, activities can be performed.

  Further, in the present embodiment, the joint motor 40 and the wheel motor 50 are controlled so that the leg wheel type robot 100 turns without moving back and forth at a predetermined turning center position (super turning). At this time, control using the coordinates of the center position on the base body 10 as the turning center coordinates (super-spinning turning control) and control using coordinates other than the center position on the base body 10 as the turning center coordinates (turning center offset type super-spinning turning Control).

  Thereby, when the center position of the base body 10 is set as the turning center position, the leg-wheel type robot can be turned with the minimum turning radius. Further, when the center position other than the center position of the base body 10 is set as the turning center position, when the center position of the base body 10 and the center of gravity position are different, the center wheel position can be used as a turning center, so that the super wheel can be turned. It is possible to turn the robot 100 in a balanced manner.

  In the second embodiment, the leg portion 12 corresponds to the leg wheel mechanism portion of the invention 8, and the joint motor 40 of the rotary joint 14 is the actuator of the invention 1, or the second invention of the inventions 2 to 5, 7 or 8. Corresponding to one actuator, the joint motor 40 of the rotary joints 16 and 18 corresponds to the second actuator of the inventions 2 to 5, 7 or 8, and the CPU 60 corresponds to the control means of the inventions 2 to 5, 7 or 8. ing.

In the first embodiment, the joint motor 40 is controlled so that the drive wheel 20 is grounded on the yaw axis at the time of turning. However, the present invention is not limited to this. First, when the base body 10 is displaced in the vertical direction, the joint motor 40 can be controlled so that the drive wheel 20 is grounded on the yaw axis.
FIG. 11 is a diagram illustrating an operation in an area where the vehicle height is limited and the ceiling is low.

As shown in FIG. 11, in an area with a low ceiling where the vehicle height is restricted, the base body is maintained while maintaining the positional relationship between the projection point of the center of gravity of the base body 10 on the floor surface and the quadrangle formed by the drive wheels 20. Since 10 can be moved up and down, an activity in an area with such a low ceiling is possible.
In the first embodiment, the leg-wheel type robot according to the present invention is applied to the case of going over the stairs. However, the present invention is not limited to this, and the same applies to the case of getting over steps other than the stairs. Can do.

  In the first and second embodiments, the configuration of the leg-wheel type robot 100 according to the present invention is such that the base 10 has a pair of four legs 12 on the front side and a pair of left and right sides on the rear side. However, the present invention is not limited to this, such as a configuration in which a pair of left and right legs 12 is provided in the center of the base 10, a configuration in which three legs 12 are provided symmetrically, a configuration in which five or more legs 12 are provided, and the like. Other configurations may be used without departing from the spirit of the invention. In the case of a multi-legged configuration, the unnecessary leg 12 may be controlled so as not to be used for traveling control.

[Third Embodiment]
Next, a third embodiment of the present invention will be described with reference to the drawings. FIGS. 16 to 34 are diagrams showing a third embodiment of the leg-wheel type robot, the object recognition apparatus, and the object recognition method according to the present invention.

First, the configuration of a leg wheel type robot 100 to which the present invention is applied will be described.
FIG. 16 is a front view of the leg wheel type robot 100.
FIG. 17 is a side view of the leg wheel type robot 100.
As shown in FIGS. 16 and 17, the leg-wheel type robot 100 includes a base body 10 and four leg portions 12 coupled to the base body 10.

Two legs 12 are connected to the front portion of the base body 10 via a rotary joint 14 at symmetrical positions. In addition, two legs 12 are connected to the rear part of the base body 10 via a rotary joint 14 at symmetrical positions. The rotary joint 14 rotates with the direction orthogonal to the bottom surface of the leg wheel type robot 100 as an axial direction. That is, it rotates around the yaw axis.
Each leg portion 12 is provided with two rotary joints 16 and 18. The rotary joint 14 rotates with the lower side as the axial direction, and the rotary joints 16 and 18 rotate with the direction orthogonal to the side surface of the leg wheel type robot 100 as the axial direction when the rotary joint 14 is in the state of FIG. . That is, when the rotary joint 14 is in the state of FIG. 16, it rotates around the pitch axis, and when the rotary joint 14 is rotated 90 degrees from the state of FIG. 16, it rotates around the roll axis. Therefore, each leg 12 has three degrees of freedom.

A driving wheel 20 is rotatably provided at the tip of each leg 12 with the same axial direction as the rotary joints 16 and 18.
At the tip of each leg 12, a front leg tip sensor 22 for measuring the distance to an object existing on the movement path of the leg wheel type robot 100 and a lower leg tip sensor 24 for measuring the distance to the ground plane are provided. Is provided.

  On the other hand, a three-dimensional distance measuring device 200 is attached to the front surface of the base body 10. The coordinate system (hereinafter referred to as a sensor coordinate system) of the three-dimensional distance measuring apparatus 200 has a depth (front-rear length) direction of the base body 10 as an xrs axis, and a width (left-right length) direction of the base body 10 as a yrs axis. The height direction of the substrate 10 is the zrs axis, the xrs axis is the front of the substrate 10, the yrs axis is the right side of the substrate 10, and the zrs axis is the positive direction above the substrate 10.

Next, the external structure of the three-dimensional distance measuring device 200 will be described.
FIG. 18 is a front view (yrs-zrs plane) of the three-dimensional distance measuring apparatus 200.
FIG. 19 is a side view (xrs-zrs plane) of the three-dimensional distance measuring apparatus 200.
FIG. 20 is a top view (yrs-xrs plane) of the three-dimensional distance measuring apparatus 200.
18 to 20, the three-dimensional distance measuring device 200 is provided above the lower support plate 204, the upper support plate 206 provided above the lower support plate 204, and above the upper support plate 206. The camera support plate 208 is configured. The lower support plate 204 and the upper support plate 206, and the upper support plate 206 and the camera support plate 208 are supported by a plurality of columns, respectively.

As shown in FIG. 19, the camera support plate 208 is provided so as to protrude in the positive direction of the xrs axis with respect to the lower support plate 204 and the upper support plate 206. A camera 222 is attached to the protruding portion of the camera support plate 208.
A motor 216 is attached to the lower surface of the lower support plate 204. A rotation shaft of the motor 216 (hereinafter referred to as a drive rotation shaft) penetrates the lower support plate 204 from below and is connected to a pulley 220 a disposed between the lower support plate 204 and the upper support plate 206. .

  On the other hand, a two-dimensional distance measuring device 212 is attached to a lower surface of the lower support plate 204 at a position spaced apart from the motor 216 in the horizontal direction by a predetermined distance. The two-dimensional distance measuring device 212 has a rotation shaft (hereinafter referred to as a driven rotation shaft), and the driven rotation shaft passes through the lower support plate 204 from below, and is between the lower support plate 204 and the upper support plate 206. The pulley 220b is connected to the pulley 220b.

A belt 221 is wound around the pulleys 220a and 220b. Therefore, when the pulley 220a is rotated by the motor 216 and the pulley 220b is rotated by the belt 221 wound around the pulley 220a, the two-dimensional distance measuring device 212 rotates around the zrs axis as shown in FIG.
FIG. 21 is a diagram illustrating a scanning range of the distance measuring sensor.

  The two-dimensional distance measuring device 212 has a built-in distance measuring sensor and, as shown in FIG. 21, rotates the distance measuring sensor around a zrs axis and an axis orthogonal to the measuring direction at every predetermined scanning unit angle. The measurement result of the distance measuring sensor is acquired. The scanning range of the distance measuring sensor is set so that the leg wheel type robot 100 scans mainly below the leg wheel type robot 100 because the leg wheel type robot 100 is intended to move up and down stairs and avoid obstacles. . At the origin position of the two-dimensional distance measuring device 212 and the distance measuring sensor (the positions where the scanning angles θ and φ are 0 °), the measuring direction of the distance measuring sensor coincides with the xrs axis, and the rotation axis of the distance measuring sensor is It matches the yrs axis. Although the direction of the rotation axis of the distance measurement sensor changes depending on the scanning angle of the two-dimensional distance measurement device 212, it coincides with the yrs axis at the origin position. Therefore, for convenience of explanation, the rotation axis of the distance measurement sensor is expressed as the yrs' axis. To do.

  In the three-dimensional distance measuring apparatus 200, the rotation driving mechanism (motor 216, encoder 218, pulleys 220a and 220b, belt 221 and lower support plate 204) that rotates and drives the two-dimensional distance measuring apparatus 212 has a scanning range shown in FIG. Since it is provided outside, within the scanning range shown in FIG. 21, the scanning of the distance measuring sensor 212 a is not hindered by each mechanism unit constituting the three-dimensional distance measuring apparatus 200.

  Further, since it is only necessary to be able to recognize an obstacle on the travel route of the leg-wheel type robot 100, the scanning range by the rotational drive around the zrs axis does not need to cover up to 180 ° forward, and the horizontal range from the two-dimensional distance measuring device 212 is not necessary. A range in which the motor 216 and the encoder 218 arranged at a predetermined distance in the direction are out of the scanning range is also sufficient. Therefore, the rotational driving range of the two-dimensional distance measuring device 212 is set to a range not including the motor 216 and the encoder 218.

Next, the movement control system of the leg wheel type robot 100 will be described.
FIG. 22 is a block diagram showing a movement control system of the leg wheel type robot 100.
As shown in FIG. 22, joint motors 40 that rotationally drive the rotary joints 14 to 18 are respectively provided in the rotary joints 14 to 18 of the leg portions 12. Each joint motor 40 is provided with an encoder 42 that detects the rotational angle position of the joint motor 40, and a driver 44 that controls the driving of the joint motor 40 based on the motor command signal and the output signal of the encoder 42.

A wheel motor 50 that rotationally drives the drive wheel 20 is provided on the drive wheel 20 of each leg 12. Each wheel motor 50 is provided with an encoder 52 that detects the rotational angle position of the wheel motor 50, and a driver 54 that controls the driving of the wheel motor 50 based on the motor command signal and the output signal of the encoder 52.
The leg-wheel type robot 100 further includes a CPU 60, a three-axis attitude sensor 70 that detects the attitude of the leg-wheel type robot 100, a wireless communication unit 74 that performs wireless communication with an external PC, the wireless communication unit 74, and the CPU 60. Are provided with a hub 76 that relays the input / output of the sound and a speaker 78 that outputs a warning sound or the like.

The triaxial attitude sensor 70 includes a gyroscope or an acceleration sensor, or both, and detects the inclination of the attitude of the leg wheel type robot 100 with respect to the ground axis.
The CPU 60 outputs motor command signals to the drivers 44 and 54 via the motor command output I / F 61 and inputs output signals of the encoders 42 and 52 via the angle fetch I / F 62. In addition, sensor signals are input from the three-dimensional distance measuring device 200, the front leg tip sensor 22, the lower leg tip sensor 24, and the three-axis posture sensor 70 via the sensor input I / F 63, respectively. Further, signals are input / output to / from the hub 76 via the communication I / F 64, and an audio signal is output to the speaker 78 via the sound output I / F 65.

Next, the control structure of the two-dimensional distance measuring device 212 will be described.
FIG. 23 is a block diagram showing a control structure of the two-dimensional distance measuring device 212.
As shown in FIG. 23, the two-dimensional distance measuring device 212 includes a distance measuring sensor 212a that measures a distance to a measurement point on an object that exists within the measurement range, a motor 212c that rotationally drives the distance measuring sensor 212a, The encoder 212d is configured to detect the rotational angle position of the motor 212c, and the driver 212b is configured to control the driving of the motor 212c based on the command signal and the output signal of the encoder 212d.

The driver 212b has a scanning angle range (for example, a predetermined angle range such as −40 ° to + 40 °) and a scanning unit angle (for example, a predetermined unit such as 0.36 °) set in the command signal from the sensing processor 210. Based on the angle, control is performed to rotate the rotation axis of the motor 212c by a scanning unit angle.
The motor 212c is provided so as to rotate and drive a laser output unit (not shown) and a light receiving unit (not shown) of the distance measuring sensor 212a around the yrs' axis. Are rotated by a scanning unit angle (Δθ).

Next, the control structure of the three-dimensional distance measuring apparatus 200 will be described.
FIG. 24 is a block diagram showing a control structure of the three-dimensional distance measuring apparatus 200.
As shown in FIG. 24, the three-dimensional distance measuring apparatus 200 includes a sensing processor 210, a two-dimensional distance measuring apparatus 212, a motor 216, an encoder 218, a command signal, and an output signal of the encoder 218. A driver 214 that controls driving and a camera 222 are included.

  The sensing processor 210 executes a dedicated program, gives a command signal to the driver 212b, rotates the distance measuring sensor 212a, and is within an area that can be measured within the scanning range of the distance measuring sensor 212a (hereinafter referred to as a scanning plane). A two-dimensional distance measuring device that executes a first scanning process for measuring a distance to a measurement point on an existing object and gives a command signal to the driver 214 each time the first scanning process for one scanning plane is completed. A second scanning process for rotating 212 is executed.

The sensing processor 210 further operates on the object existing in the measurement range based on the distance information (hereinafter referred to as distance information) measured by the two-dimensional distance measurement device 212 through the first scanning process and the second scanning process. The process of recognizing the continuous surface is executed.
Next, the principle of distance measurement of the three-dimensional distance measuring apparatus 200 will be described.
FIG. 25 is a diagram for explaining the principle of distance measurement of the two-dimensional distance measuring device 212.

  As shown in FIG. 25, the two-dimensional distance measuring device 212 is rotated each time the distance measuring sensor 212a rotates by the scanning unit angle around the yrs' axis according to the rotational drive of the rotating shaft of the motor 212c. In addition, laser light is output from the laser output unit, and reflected light from an object (obstacle in FIG. 25) with respect to the output light is received by the light receiving unit, and a distance corresponding to each scanning angle (measurement distance in FIG. 25). L (distance between the object and the light receiving unit)) is measured.

FIG. 26 is a diagram illustrating a case where scanning is performed by the first scanning process and the second scanning process. FIG. 6A is a diagram showing the relationship between the measurement distance L and the scanning angle θ when the distance measurement sensor 212a is rotated about the yrs ′ axis, and FIG. 6B is a two-dimensional distance measurement device. It is a figure which shows the relationship between the scanning plane when 212 is rotated around the zrs axis, and the scanning angle φ.
In the first scanning process, for example, as shown in FIG. 26A, the distance measuring sensor 212a is rotated by the scanning unit angle around the yrs' axis while each scanning angle (in FIG. This is a process of measuring distance information (L (θ ₁ ), L (θ ₂ ), L (θ ₃ ) in FIG. 26A) according to θ ₁ , θ ₂ , θ ₃ ).

In the first scanning process, a plane formed by connecting the rotation center of the rotating shaft of the motor 212c and both ends of the scanning trajectory line of the laser is a scanning plane (a fan-shaped plane when no object is present). .
The driver 214 performs control to rotate the rotation axis of the motor 216 by the scanning unit angle based on the scanning angle range and the scanning unit angle (Δφ) set in the command signal from the sensing processor 210.

  The motor 216 is provided to rotationally drive the two-dimensional distance measuring device 212 around the zrs axis via a speed reducer (not shown), pulleys 220a and 220b, and a belt 221, and receives a control signal from the driver 214. Correspondingly, the rotation axis of the self is rotated by the scanning unit angle. As a result, the rotational driving force is transmitted to the driven rotational shaft via the pulleys 220a and 220b in accordance with the rotational driving of the rotational shaft of the motor 216, and the two-dimensional distance measuring device 212 rotates about the zrs axis by a scanning unit angle. .

That is, the second scanning process is a process of rotating the two-dimensional distance measuring device 212 by the scanning unit angle around the zrs axis as shown in FIG. Then, by performing the first scanning process and the second scanning process alternately and continuously, the scanning plane formed by the first scanning process is continuously formed around the zrs axis.
FIG. 27 is a diagram illustrating a distance measurement example of the three-dimensional distance measurement apparatus 200.

Thereby, for example, as shown in FIG. 27, the three-dimensional shape of an object such as a wall, a wall, a stand, or a shelf can be grasped.
Further, as shown in FIG. 26B, the distance information of each measurement point after the second scanning process is expressed as L (θ _i , φ _j ). Here, i is a serial number given to each measurement point according to the scanning angle around the yrs' axis, and j is a serial number given to each measurement point according to the scanning angle around the zrs axis.

Next, an object recognition process executed by the three-dimensional distance measuring apparatus 200 will be described.
FIG. 28 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200.
The object recognition process is a process realized by the sensing processor 210 reading a dedicated program stored in a ROM (not shown) based on a command signal from the CPU 60 and executing the read program. When the process is executed, as shown in FIG. 28, first, the process proceeds to step S300.

In step S300, in the three-dimensional distance measuring apparatus 200, the scanning angle range and the scanning unit angle of the distance measuring sensor 212a and the two-dimensional distance measuring apparatus 212 are set based on the command signal from the CPU 60, and the process proceeds to step S302. Here, the command signal from the CPU 60 includes information on the scanning angle range and the scanning unit angle.
In step S302, by outputting a command signal to the two-dimensional distance measuring device 212, the driver 212b, the motor 212c, and the encoder 212d are driven, and the distance measuring sensor 212a is scanned around the yrs' axis set in step S300. Within the range, the scanning unit angle set in step S300 is rotated about the yrs' axis, and the first scanning process for measuring the distance information corresponding to each scanning angle is executed, and the process proceeds to step S304.

In step S304, a filtering process using a median filter is performed on the distance information measured in step S302 to remove noise components, and the process proceeds to step S306.
In step S306, the distance information of the rotating coordinate system after noise removal in step S304 is converted into the coordinate information of the orthogonal coordinate system. Thereby, the distance information of each measurement point obtained by the first scanning process is converted into coordinate information of an orthogonal coordinate system (hereinafter referred to as a scanning plane coordinate system) in which the scanning plane of the first scanning process is a two-dimensional plane. Converted.

Next, the process proceeds to step S310, and a line segment in the orthogonal coordinate system is detected by Hough transform based on the coordinate information transformed in step S306.
FIG. 29 is a diagram for explaining the principle of the Hough transform. FIG. 4A shows the xy plane, and FIG. 4B shows the ρ-θ plane. Note that the x-axis and y-axis in FIG. 5A indicate axes in the scanning plane coordinate system and are separate from the axes in the sensor coordinate system.

  The Hough transform is one of feature extraction methods used in digital image processing. Classically, it used to detect straight lines, but it is more generalized and used for various forms (those that can be expressed in the form of equations such as circles and ellipses). A feature of the Hough transform is that a relatively good result can be obtained even when a straight line in the image is cut off in the middle or when noise exists.

Consider a straight line ax + by + c = 0 in the scanning plane coordinate system, as shown in FIG.
When a perpendicular is drawn from this straight line to the origin, the length of the perpendicular is ρ, and the angle between the perpendicular and the x axis is θ, this straight line can be expressed by the following equation (15).

The above equation (15) can be transformed into the following equation (16).

ρ = xcosθ + ysinθ (16)

Therefore, one straight line corresponds to one set of (ρ, θ). Here, the point (ρ, θ) is referred to as the Hough transform of the straight line ax + by + c = 0. A straight line group passing through an arbitrary point (x0, y0) in the scanning plane coordinate system can be expressed by the following equation (17).

ρ = x0cosθ + y0sinθ (17)

Here, when the locus of each straight line group passing through the three points P1, P2, and P3 on the xy plane is drawn on the ρ-θ plane, a sinusoidal curve is obtained as shown in FIG. If these three points exist on the same line in the xy plane, the values of ρ and θ are the same, and the curves corresponding to the three points intersect at one point on the ρ-θ plane.

If the principle of the Hough transform is used, a line segment can be detected based on the coordinates of a plurality of measurement points. That is, for n (n ≧ 2) measurement points, n curves are drawn on the ρ-θ plane, and m (n ≧ m ≧ 2) curves among them intersect at one point. For example, the m measurement points corresponding to the m curves are on the same straight line in the xy plane.
Next, as shown in FIG. 28, the process proceeds to step S320, and the end point of the detected line segment is determined as the boundary (continuous edge) of the continuous surface. When multiple line segments overlap or when multiple line segments exist within a predetermined distance, it is considered as one line segment, and the two farthest points among the end points of these line segments are Judge as a boundary.

Next, the process proceeds to step S324, where the coordinate information of the end point determined as the boundary of the continuous surface is converted into a sensor coordinate system, and the converted coordinate information is associated with each line segment in a memory (not shown) such as a RAM. Store, and the process proceeds to step S326.
In step S326, it is determined whether or not the processing in steps S302 to S324 has been completed for all scanning planes corresponding to the scanning angle range and scanning unit angle of the second scanning processing, and when it is determined that the processing has been completed (Yes) ) Proceeds to step S336.

  In step S336, surface data is generated based on the coordinate information stored in the memory. Since the line segment connecting the end points determined as the boundary of the continuous plane (which is regarded as one line segment in step S320) is an intersection line where the continuous plane and the scanning plane intersect, generation of plane data is, for example, The slope and coordinates of the line segment connecting the end points determined as the boundary of the continuous surface in a certain scanning plane and the line segment connecting the end points determined as the boundary of the continuous surface in the scanning plane adjacent to the zrs axis are within a predetermined range. Are determined as continuous surfaces, coordinate information corresponding to these line segments is associated, or connected by using a known interpolation method. For example, since a continuous surface having an inclination close to 0 can be regarded as a horizontal surface, it can be determined that the surface is a walking surface.

Next, the process proceeds to step S338, and the surface data generated in step S336 is output to the CPU 60 via the hub 76 and the communication I / F 64, and the series of processing ends.
On the other hand, when it is determined in step S326 that the processes in steps S302 to S324 are not completed for all scanning planes (No), the process proceeds to step S340, and a command signal is output to the three-dimensional distance measuring apparatus 200. The driver 214, the motor 216, and the encoder 218 are driven, and the two-dimensional distance measuring device 212 is rotated around the zrs axis by the scanning unit angle set in step S300 within the scanning angle range around the zrs axis set in step S300. The second scanning process is performed to rotate to step S302, and the process proceeds to step S302.

Next, processing executed by the CPU 60 will be described.
The CPU 60 activates a control program stored in a predetermined area such as a ROM, and executes the elevation control process shown in the flowchart of FIG. 30 according to the control program.
FIG. 30 is a flowchart showing the elevation control process.
The elevation control process is a process for performing the elevation control of the leg portion 12. When the elevation control process is executed by the CPU 60, first, the process proceeds to step S400 as shown in FIG.

In step S400, surface data is input from the three-dimensional distance measuring apparatus 200, the process proceeds to step S402, and the coordinates of each measurement point in the sensor coordinate system are converted into coordinates in the global coordinate system based on the input surface data. A point on the periphery of the continuous surface is detected as a feature point of the staircase.
Next, the process proceeds to step S404, the width of the staircase is calculated based on the detected feature point of the staircase, and the process proceeds to step S406, where the actual coordinates of the stair nosing part of the staircase are calculated based on the detected feature point of the staircase. Then, the process proceeds to step S408.

In step S408, inverse kinematics calculation and centroid calculation are performed based on the calculated width of the staircase and the actual coordinates of the stair nose and the sensor signal of the three-axis posture sensor 70, the process proceeds to step S410, and the calculation result of step S408 The landing position of the leg tip (drive wheel 20) is determined based on the above, and the process proceeds to step S412.
In step S412, sensor signals are input from the front leg tip sensor 22 and the lower leg tip sensor 24, respectively, and the process proceeds to step S414 to calculate the distance to the kick plate based on the input sensor signal of the front leg tip sensor 22. Then, the process proceeds to step S416, the positional relationship between the leg tip and the tread is calculated based on the input sensor signal of the lower leg tip sensor 24, and the process proceeds to step S418.

In step S418, a motor command signal to the drivers 44 and 54 is generated based on the determined landing position and the calculated both distances, the process proceeds to step S420, and the generated motor command signal is output to the drivers 44 and 54. The process proceeds to step S422.
In step S422, it is determined whether or not the leg tip has landed on the tread. When it is determined that the leg tip has landed (Yes), the series of processes is terminated and the original process is restored.

On the other hand, when it is determined in step S422 that the leg tip does not land (No), the process proceeds to step S412.
Next, the operation of the present embodiment will be described.
A case will be described in which a stairway exists on the movement path of the leg-wheel type robot 100 and the stairs are overcome.

In the sensing processor 210, first, through step S300, the scanning angle range and the scanning unit angle are set in the sensing processor 210 based on the command signal from the CPU 60.
Here, it is assumed that the two-dimensional distance measuring device 212 is a two-dimensional range sensor having a distance measurement range of 20 to 4095 [mm], a maximum scanning angle range of 240 °, and an angular resolution of 0.36 °. If the scanning angle range is 240 °, it is the same as the scanning angle range forming the scanning plane shown in FIG. 21, and the lower support plate 204, the pulleys 220a and 220b, and the belt 221 above the two-dimensional distance measuring device 212 are provided. It will not be included in the scanning range.

  Further, a scanning angle range of 240 ° and a scanning unit angle of 0.36 ° are set for the first scanning process, and a scanning angle range of −40 ° to + 40 ° and a scanning unit angle of 10 ° are set for the second scanning process. Is set (in this case, nine scanning planes are formed). If the scanning angle range is −40 ° to + 40 °, the motor 216 and the encoder 218 arranged at a predetermined distance in the horizontal direction from the two-dimensional distance measuring device 212 are not included in the scanning range.

Next, through step S302, a command signal is output to the driver 212b based on the scan angle range and scan unit angle set for the first scan process, whereby the first scan process is executed. Initially, a first scanning process is performed on a position (origin position) where the scanning angle φ around the zrs axis is 0 °.
As a result, based on the command signal from the sensing processor 210 and the output signal from the encoder 212d, the driver 212b drives the rotation shaft of the motor 212c to rotate, and the distance measuring sensor 212a scan unit angle Δθ = about the yrs ′ axis. While rotating by 0.36 °, the distance corresponding to each scanning angle is measured. Each distance information is output to the sensing processor 210 as a data string L (θ _i , φ _j ).

In the scanning range around the yrs' axis, there is no object other than the base 10 that obstructs scanning, such as the driving mechanism of the two-dimensional distance measuring device 212. Distance information can be obtained.
When the first scanning process for one scanning plane is completed, the filtering process is performed on the distance information measured in the first scanning process through step S304. Thereby, the noise component in measurement information is removed.

FIG. 31 is a graph showing a measurement result by the first scanning process.
Here, the measurement distance L [mm] of each measurement point after removing the noise component is as shown in FIG. 31, for example. In FIG. 31, the horizontal axis is the number of the measurement point corresponding to each scanning angle (first scanning angle number), and the vertical axis is the measurement distance L [mm] to the measurement point of each scanning angle number.
In the example of FIG. 31, it can be seen that there are stairs on the walking path of the leg-wheel type robot 100 with a constant step and a tread board as a continuous surface.

Next, through step S306, the distance information of the rotated coordinate system after the filtering process is converted into the coordinate information of the orthogonal coordinate system.
FIG. 32 is a graph showing the result of converting the distance information of the rotating coordinate system of each measurement point of FIG. 31 into the coordinate information of the orthogonal coordinate system.
The distance information of the rotational coordinate system of each measurement point in FIG. 31 is obtained by coordinate conversion, as shown in FIG. 32, the distance [mm] in the xrs axis direction and the distance [mm] in the zrs axis direction corresponding to each scanning angle. Is two-dimensional coordinate information. In FIG. 32, the horizontal axis represents the distance L _x [mm] in the xrs axis direction, and the vertical axis represents the distance L _z [mm] in the zrs axis direction.

Next, through step S310, a line segment in the orthogonal coordinate system is detected by Hough transform based on the transformed coordinate information.
FIG. 33 is a graph showing measurement points in the orthogonal coordinate system and the results of the Hough transform.
In the orthogonal coordinate system, each measurement point is represented as a set of a plurality of points along the contours of the kick board and the tread board as shown in FIG. In the example of FIG. 33 (a), in the region corresponding to the first step tread, some measurement points are distributed in a region deviating from the continuous surface. The measurement results vary due to the influence of the error area A1. The area A2 has a lower measurement resolution than the other areas.

When the Hough transform is performed on the measurement result, as shown in FIG. 33B, a line segment along the contour of each tread is detected.
As another method for detecting a line segment based on a plurality of measurement points, a least square method is known.
However, in the least square method, line segments are detected by tracing each measurement point, and therefore, in the low resolution area A2, a line segment that does not follow the contour of the tread may be detected. On the other hand, in the Hough transform, it is difficult to be influenced by the measurement resolution, and even if the low resolution region A2 is included, as shown in FIG. Can do.

  Further, in the least square method, line segments are detected by tracing each measurement point, so that the error region A1 is detected as an oblique line segment that is actually a flat line segment. On the other hand, in the Hough transform, if it is not easily affected by variations and the number of variations is small, the error region A1 and the regions on both sides thereof can be detected as flat line segments as shown in FIG. it can.

  Further, the least square method has a problem as to whether to recognize a region from where to where is one continuous surface. In this case, for example, it is conceivable to recognize a point where the slope of the detected line segment changes abruptly as the boundary of the continuous surface. However, in this recognition method, the error region A1 is actually the error region A1 and the regions on both sides thereof. Is a single continuous surface, the regions on both sides are recognized as separate continuous surfaces. On the other hand, in the Hough transform, if it is not easily affected by variations and the number of variations is small, the error region A1 and the regions on both sides thereof are detected as one flat line segment as shown in FIG. be able to.

Next, through steps S320 and S324, the end point of the detected line segment is determined as the boundary of the continuous surface, the coordinate information of the end point determined as the boundary of the continuous surface is converted into the sensor coordinate system, and the converted coordinate information Is stored in the memory.
When the measurement is completed for one scanning plane, the second scanning is performed by outputting a command signal to the driver 214 based on the scanning angle range and the scanning unit angle set for the second scanning process through step S340. Processing is executed.

  As a result, based on the command signal from the sensing processor 210 and the output signal from the encoder 218, the driver 214 rotates the rotation axis of the motor 216, and the two-dimensional distance measurement device 212 has a scanning unit angle of 10 around the zrs axis. Rotate by °. By the second scanning process, the orientation of the two-dimensional distance measuring device 212 changes by 10 ° around the zrs axis with respect to the previous state. In this state, the first scanning process is executed again through step S302. That is, a new scanning plane is formed at a position shifted by 10 ° around the zrs axis, and the first scanning process is performed on this scanning plane.

Here, since the two-dimensional distance measuring device 212 is attached to the front surface of the base body 10, the base body 10 is included in the scanning range, but in the range including the front of the leg wheel type robot 100 and the walking path. Since there are no obstructions, it can be said that a sufficient scanning range for performing the walking control of the leg-wheel type robot 100 can be secured.
When measurement is completed for all scanning planes, surface data is generated based on the coordinate information stored in the memory through step S336.

FIG. 34 is a diagram illustrating determination results of continuous surfaces.
As shown in FIG. 34, the plane data is generated by connecting line segments having similar inclinations and coordinates between adjacent scan planes around the zrs axis. In the example of FIG. 34, for example, the line segment of the region (φ0, 2) corresponding to the first step board on the scanning plane φ0 and the region (φ1, 2) corresponding to the first step plate on the scanning plane φ1. ) Is determined to constitute one continuous surface, surface data in which the coordinate information of the end points of the line segments is associated is generated for the continuous surface.

  Similarly, the line segment of the region (φ0, 2i (i is an integer greater than or equal to 2)) in the scan plane φ0 and the line segment of the region (φ1, 2i) in the scan plane φ1 represent one continuous plane, and the scan plane φ0 It is determined that the line segment of the region (φ0, 2j (j is an integer equal to or greater than 1) -1) and the line segment of the region (φ1, 2j-1) on the scanning plane φ1 constitute one continuous surface, and the plane data is Generated.

The surface data is output to the CPU 60 via the hub 76 and the communication I / F 64 through step S338.
When the surface data is input, the CPU 60 detects the staircase feature point based on the input surface data through step S402. In addition, the input surface data is analyzed, and for example, a continuous surface whose inclination is close to 0 is regarded as a horizontal surface, and it is determined that the leg-wheel type robot 100 can walk.

If it is determined that the surface is a walkable surface, the width of the staircase and the actual coordinates of the stair nose are calculated based on the detected feature points of the staircase through steps S404 to S410. The landing position of the leg tip is determined based on the real coordinates of the part.
Further, through steps S412 to S416, sensor signals are input from the leg tip sensors 22 and 24, respectively, and the distance to the kick plate and the positional relationship between the leg tip and the tread plate are calculated. Then, through steps S418 and S420, a motor command signal is generated based on the determined landing position and the calculated both distances, and the generated motor command signal is output to the drivers 44 and 54. As a result, the driving wheel 20 rotates and the rotary joints 14 to 18 are driven, and the leg-wheel type robot 100 gets over the stairs while keeping its posture properly. Depending on the situation, the stairs are avoided and stopped. Therefore, the adaptability to the stairs is high like the legged robot.

In addition, although the step is given as an example, the staircase where the step board is a continuous surface is given as an example, but the step can be accurately recognized even for a staircase where the step is not constant, a staircase without a kick plate, etc. The adaptability to the stairs of the leg wheel type robot 100 can be improved.
On the other hand, on a flat ground, the leg-wheel type robot 100 can move by wheel running. Therefore, the mobility on the flat ground is high like the wheel type robot.

  As described above, in this embodiment, the distance measuring sensor 212a that measures the distance to the measurement point on the object is provided, the distance measurement sensor 212a is scanned, and the distance measurement sensor can be measured for the measurement points that can be measured in the scanning range. The measurement result of 212a is acquired, the acquired measurement result is converted into the coordinates of the orthogonal coordinate system, and the line segment in the orthogonal coordinate system is detected by the Hough transform based on the converted coordinates of the at least two measurement points, and detected. Recognize continuous faces or boundaries of continuous faces based on line segments.

  Accordingly, since at least the planar shape of the object can be grasped as a continuous surface or a boundary between continuous surfaces, the posture control of a robot that requires complicated posture control, such as the legged robot or the legged wheel type robot 100, is possible. A suitable recognition result can be obtained. Further, since the line segment is detected by the Hough transform, when the object recognition is performed by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a, there is a possibility that the recognition accuracy may be lowered due to a decrease in measurement resolution or a variation in measurement results. Can be reduced.

Furthermore, as compared with the least square method, robust detection can be performed for an edge that is difficult to detect.
Furthermore, in this embodiment, the continuous surface or the boundary of the continuous surface is recognized based on the coordinates of the end points of the detected line segment.
Thereby, the continuous surface or the boundary of the continuous surface can be recognized relatively accurately.

  Further, in the present embodiment, a yrs′-axis rotation mechanism including a motor 212c and the like that rotates the ranging sensor 212a around the yrs ′ axis, and a zrs axis including a motor 216 and the like that rotates the ranging sensor 212a around the zrs axis. A first scanning for obtaining a measurement result of the distance measuring sensor 212a for each scanning unit angle of the yrs 'axis rotating mechanism while rotating the distance measuring sensor 212a by the yrs' axis rotating mechanism. By performing the second scanning performed for each scanning unit angle of the zrs axis rotation mechanism while rotating the distance measuring sensor 212a, the measurement is performed for each scanning unit angle of the yrs' axis rotation mechanism and for each scanning unit angle of the zrs axis rotation mechanism. Get the result.

  As a result, the three-dimensional shape of the object can be grasped as a continuous surface or a boundary between the continuous surfaces, so that the posture control of a robot that requires complicated posture control, such as the legged robot or the leg-wheel type robot 100, is possible. A more suitable recognition result can be obtained. In addition, since a rotation mechanism that rotates the distance measuring sensor 212a is adopted, a space required for scanning is smaller than that of the moving mechanism, the scanning mechanism is simplified, and high-speed scanning is realized. Can do.

Furthermore, in this embodiment, the motor 216 and the encoder 218 and the two-dimensional distance measuring device 212 are arranged at a predetermined distance in the horizontal direction, and the rotational driving force of the driving rotary shaft is set to the pulley 220a, the belt 221 and the pulley 220b. The two-dimensional distance measuring device 212 is driven to rotate about the zrs axis.
As a result, there is no obstacle to scanning within the scanning angle range of the two-dimensional distance measuring device 212, so that accurate distance information can be obtained. Further, since the motor 216 and the encoder 218 and the two-dimensional distance measuring device 212 are arranged in the horizontal direction, the occupation ratio in the height direction of the three-dimensional distance measuring device 200 can be reduced.

Further, in the present embodiment, leg tip sensors 22 and 24 are provided, the stairs are recognized based on the distance measured by the leg tip sensors 22 and 24, and the motors 40 and 50 are controlled based on the recognition result.
Thereby, since the raising / lowering control of the leg part 12 is performed while recognizing the unknown staircase using the leg tip sensors 22 and 24, it is possible to realize higher adaptability to the unknown staircase than in the past. . In addition, since it can operate in an environment where people are active, it is suitable for home robots, personal robots, and the like that are used for acting with people.

Further, in the present embodiment, the three-dimensional distance measuring device 200 is provided on the front surface of the base body 10, and the leg tip sensors 22 and 24 are provided on the distal ends of the leg portions 12.
As a result, it is possible to detect an object existing on the movement path of the leg wheel type robot 100 with a wide field of view, and to accurately measure the distance between the drive wheel 20 and the staircase when moving up and down the stairs.

Further, in the present embodiment, the distance to the stair riser plate is calculated based on the measurement result of the front leg tip sensor 22, and the positions of the drive wheels 20 and the step board of the staircase are calculated based on the measurement result of the lower leg tip sensor 24. Calculate the relationship.
Thereby, since the characteristic more effective for the raising / lowering control of the leg part 12 can be detected among the characteristics of the staircase, higher adaptability can be realized for the unknown staircase.

  In the third embodiment, steps S302 and S340 correspond to the measurement result acquisition unit of the invention 9, 12 or 14, or the measurement result acquisition step of the invention 25, and step S306 includes the coordinate conversion unit of the invention 9, Or it corresponds to the coordinate conversion step of the invention 25. Step S310 corresponds to the line segment detection means of the invention 9, 11 or 13 or the line segment detection step of the invention 25, and steps S320 and S336 correspond to the recognition means of the invention 9, 11 or 13, or the invention 25. It corresponds to the recognition step.

  In the third embodiment, the yrs 'axis corresponds to the first scanning axis of the invention 14, the zrs axis corresponds to the second scanning axis of the invention 14, and the yrs' axis rotation mechanism is the invention. The zrs axis rotating mechanism corresponds to the second scanning means of the twelfth or fourteenth aspect of the invention, or the second rotating means of the fourteenth aspect of the invention. .

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described with reference to the drawings. 35 and 36 are diagrams showing a fourth embodiment of the leg-wheel robot, object recognition apparatus, and object recognition method according to the present invention.

This embodiment differs from the third embodiment in that the Hough transform is performed after connecting the measurement points with line segments in the orthogonal coordinate system. Hereinafter, only the parts different from the third embodiment will be described, and the same parts as those in the third embodiment will be denoted by the same reference numerals and the description thereof will be omitted.
First, an object recognition process executed by the three-dimensional distance measuring apparatus 200 will be described.

FIG. 35 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200.
When the object recognition process is executed in the sensing processor 210, the process proceeds to step S308 through steps S300 to S306 as shown in FIG.
In step S308, based on the coordinate information converted in step S306, measurement points adjacent in the x-axis direction in the scanning plane coordinate system are connected with line segments, and the process proceeds to step S310 and the obtained line is changed. A line segment in the orthogonal coordinate system is detected by Hough transform based on the coordinate information of the point.

Then, after steps S320 and S324, the process proceeds to step S326, where it is determined whether or not the processing of steps S302 to S324 has been completed for all the scanning planes, and when it is determined that the processing has been completed (Yes) Through steps S336 and S338, a series of processing is completed.
On the other hand, when it is determined in step S326 that the processes in steps S302 to S324 are not completed for all scanning planes (No), the process proceeds to step S302 through step S340.

Next, the operation of the present embodiment will be described.
A case will be described in which a stairway exists on the movement path of the leg-wheel type robot 100 and the stairs are overcome. Since the third embodiment is different from the third embodiment in that a line segment is detected by Hough transform, only different portions will be described.
Through steps S308 and S310, measurement points are connected by line segments based on the converted coordinate information, and line segments in the orthogonal coordinate system are detected by Hough transform based on the coordinate information of the points on the obtained line. .

FIG. 36 is a graph showing measurement points, line segment connection results, and Hough transform results in an orthogonal coordinate system.
In the orthogonal coordinate system, each measurement point is represented as a set of a plurality of points along the contours of the kick board and the tread board, as shown in FIG. In the example of FIG. 36 (a), in the region corresponding to the first step tread, some measurement points are distributed in a region deviating from the continuous surface. The measurement results vary due to the influence of the error area A1. The area A2 has a lower measurement resolution than the other areas.

When a line segment connection is made for this measurement result, each measurement point is connected by one line as shown in FIG. As a result, between the measurement points where no measurement point exists, the points on the line are interpolated.
When the Hough transform is performed on the connection result, as shown in FIG. 36 (c), a line segment along the outline of each kick board and each tread board is detected.

If the Hough transform is performed on the measurement result in FIG. 36A without connecting the line segments, as shown in FIG. 33B, line segments along the contours of the treads are detected. The line segment along each kick plate cannot be detected.
As described above, in this embodiment, the distance measuring sensor 212a that measures the distance to the measurement point on the object is provided, the distance measurement sensor 212a is scanned, and the distance measurement sensor can be measured for the measurement points that can be measured in the scanning range. The measurement result of 212a is acquired, the acquired measurement result is converted into coordinates in an orthogonal coordinate system, the converted measurement points are connected by line segments, and orthogonality is obtained by Hough transform based on the coordinates of the points on the obtained line A line segment in the coordinate system is detected, and a continuous surface or a boundary of the continuous surface is recognized based on the detected line segment.

  Accordingly, since at least a planar shape of the object can be grasped as a continuous surface on the object or a boundary between the continuous surfaces, a robot that requires complicated posture control, such as the legged robot or the leg wheel type robot 100. A recognition result suitable for posture control can be obtained. Further, since the line segment is detected by the Hough transform, when the object recognition is performed by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a, there is a possibility that the recognition accuracy may be lowered due to a decrease in measurement resolution or a variation in measurement results. Can be reduced. In addition, since line segments are detected based on the coordinates of points on the line obtained by connecting the measurement points with line segments, relatively accurate recognition results can be obtained even if measurement resolution decreases or measurement results vary. The possibility that the recognition accuracy is lowered can be further reduced.

Furthermore, as compared with the least square method, robust detection can be performed for an edge that is difficult to detect.
In the fourth embodiment, steps S302 and S332 correspond to the measurement result acquisition means of the invention 10 or the measurement result acquisition step of the invention 26, and the step S306 corresponds to the coordinate conversion means of the invention 10 or the invention 26. Corresponding to the coordinate conversion step, step S310 corresponds to the line segment detection means of the invention 10 or the line segment detection step of the invention 26. Steps S320 and S328 correspond to the recognition means of Invention 10 or the recognition step of Invention 26.

[Fifth Embodiment]
Next, a fifth embodiment of the present invention will be described with reference to the drawings. FIGS. 37 to 44 are views showing a fifth embodiment of the leg-wheel type robot, the object recognition apparatus, and the object recognition method according to the present invention.

  The present embodiment is different from the third embodiment in that the boundary of the continuous surface is recognized based on the measurement result of the distance measuring sensor 212a and an image acquired from the camera 222 (hereinafter referred to as a camera image). Is different. Hereinafter, only the parts different from the third embodiment will be described, and the same parts as those in the third embodiment will be denoted by the same reference numerals and the description thereof will be omitted.

First, the external structure of the three-dimensional distance measuring apparatus 200 will be described.
The camera 222 faces the negative direction of the zrs axis, and can capture images of the feet and surroundings of the leg wheel type robot 100. As in the sensor coordinate system, the coordinate system of the camera 222 (hereinafter referred to as the camera coordinate system) has the same direction as the xrs axis, yrs axis, and zrs axis as the xca axis, yca axis, and zca axis, respectively. Have different origin positions. That is, the camera 222 is arranged so that the xrs-yrs plane in the sensor coordinate system and the xca-yca plane in the camera coordinate system are parallel. The xrs-zrs plane and the xca-zca plane, and the yrs-zrs plane and the yca-zca plane are also in a parallel relationship.

Next, the control structure of the three-dimensional distance measuring apparatus 200 will be described.
The sensing processor 210 further executes processing for recognizing the boundary of the continuous surface based on the measurement result of the distance measuring sensor 212a and the camera image.
Next, the principle of object recognition using the camera 222 and the three-dimensional distance measuring device 200 will be described.

FIG. 37 is a diagram for explaining the principle of object recognition using the camera 222 and the three-dimensional distance measuring apparatus 200.
First, as shown in FIG. 37, the camera 222 captures images of the feet of the leg-wheel robot 100 and its surroundings, and detects line segments from the camera image.
Next, the boundary of the continuous surface is determined as a sensor feature point from the coordinates of the measurement point measured for each scanning plane by the three-dimensional distance measuring apparatus 200, and the sensor feature point is projected onto the camera image by coordinate conversion.

And the boundary of a continuous surface is recognized based on the positional relationship of the line segment detected from the camera image (henceforth a line segment of a camera image) and the projection point of a sensor feature point.
Accordingly, the recognition accuracy can be improved as compared with the case where the recognition is performed based only on the sensor feature points obtained by the three-dimensional distance measuring apparatus 200 and the case where the recognition is performed based only on the camera image.

Next, an object recognition process executed by the three-dimensional distance measuring apparatus 200 will be described.
FIG. 38 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200.
When the object recognition process is executed in the sensing processor 210, the process first proceeds to step S250 as shown in FIG.

  In step S250, an image is acquired from the camera 222, the process proceeds to step S252, distortion correction processing is performed on the camera image, and the process proceeds to step S254, where the distortion-corrected camera image is subjected to Sobel, Canny filter, and the like. In step S256, an edge extraction process for extracting an edge is executed. In step S256, a line segment is detected from the edge-extracted camera image by Hough transform.

Subsequently, the process proceeds to step S324 through steps S300 to S320, and the coordinate information of the end point determined as the boundary of the continuous surface is converted into the sensor coordinate system, and the converted coordinate information is stored in the memory, and the process proceeds to step S326. To do.
In step S326, it is determined whether or not the processing in steps S302 to S324 has been completed for all scanning planes corresponding to the scanning angle range and scanning unit angle of the second scanning processing, and when it is determined that the processing has been completed (Yes) ) Proceeds to step S328.

In step S328, based on the coordinate information stored in the memory, the end point determined as the boundary of the continuous surface is used as the sensor feature point, and the sensor feature point is projected onto the camera image by coordinate conversion.
FIG. 39 is a diagram illustrating a sensor coordinate system, a camera coordinate system, and a coordinate system of the image sensor surface of the camera 222.

Sensor feature points can be projected onto a camera image using a general pinhole model.
The coordinate system of the image sensor surface of the camera 222 is Σ cm, the relative position between the distance measuring sensor 212a and the camera 222 is r = (rx, ry, rz), the focal length of the lens of the camera 222 is f, and the effective pixel width of the image sensor Is W and the effective pixel height of the image sensor is H, P (xrs1, yrs1, zrs1) rs = (xrs1-rx, yrs1-ry, zrs1-rz) ca in the sensor coordinate system is shown in FIG. As shown in (b), the image can be projected onto a coordinate P ′ in the coordinate system of the imaging element surface. The coordinate P ′ can be obtained by the following equation (18).

  In FIG. 39 and the above equation (18), the coordinates (a, b, c) in the sensor coordinate system Σrs, the camera coordinate system Σca, and the coordinate system Σcm of the image sensor surface are respectively represented by (a, b, c) rs, (A, b, c) ca, (a, b, c) cm.

  Next, as shown in FIG. 38, the process proceeds to step S330, and in the camera image, for each line segment of the camera image, the line segment and the sensor feature points existing within a predetermined distance from the straight line obtained by extending the line segment. And the process moves to step S332 to group the line segments associated with the same sensor feature point, and the process moves to step S334 to determine the continuous surface based on the coordinates of the end points of the line segments belonging to the same group. Determine the boundary line.

FIG. 40 is a diagram for explaining a case in which line segments of camera images and sensor feature points are associated, grouped, and boundary lines are determined.
As shown in FIG. 40A, it is assumed that two line segments l1 and l2 having close inclinations are detected from the camera image, and four projection points P1 to P4 are obtained around the line segments l1 and l2. In this case, since projection points P2, P3, and P4 exist within a predetermined distance from a straight line obtained by extending the line segment l1, the projection points P2, P3, and P4 with respect to the line segment l1 as shown in FIG. Are associated. Further, since the projection points P1, P2, and P3 exist within a predetermined distance from the straight line obtained by extending the line segment l2, the projection points P1, P2, and P3 are associated with the line segment l2.

Next, since the projection points P2 and P3 are associated in common with the line segments l1 and l2, the line segments l1 and l2 are grouped as one group.
Then, based on the coordinates of the end points P5 and P6 of the line segment l1 and the end points P7 and P8 of the line segment l2 belonging to the same group, as shown in FIG. The determined straight line is determined as the boundary line of the continuous surface.

As a result, due to the influence of contrast or the like, even though a plurality of line segments should be detected from the camera image as actually one line segment, a plurality of line segments are set to 1 based on the correspondence with sensor feature points. It can be grouped as one line segment and the boundary line of the continuous surface can be determined.
Next, as shown in FIG. 38, the process proceeds to step S337, where boundary data is generated based on the information regarding the straight line determined as the boundary line of the continuous surface, and the process proceeds to step S339, where the hub 76 and the communication I / F 64 are processed. Then, the generated boundary data is output to the CPU 60, and the series of processes is terminated.

On the other hand, when it is determined in step S326 that the processes in steps S302 to S324 are not completed for all scanning planes (No), the process proceeds to step S302 through step S340.
Next, the elevation control process will be described.
FIG. 41 is a flowchart showing the elevation control process.

When the elevation control process is executed by the CPU 60, first, the process proceeds to step S401 as shown in FIG.
In step S401, boundary data is input from the three-dimensional distance measuring apparatus 200, and the process proceeds to step S402. Based on the input boundary data, the coordinates of the camera coordinate system are converted to the coordinates of the global coordinate system, and A point on the boundary line is detected as a feature point of the staircase.

Next, after steps S402 to S420, the process proceeds to step S422, where it is determined whether the leg tip has landed on the tread. If it is determined that the leg tip has landed (Yes), the series of processing is terminated. To return to the original process.
On the other hand, when it is determined in step S422 that the leg tip does not land (No), the process proceeds to step S412.

Next, the operation of the present embodiment will be described.
A case will be described in which a stairway exists on the movement path of the leg-wheel type robot 100 and the stairs are overcome. Since the third embodiment is different from the third embodiment in that the boundary of the continuous surface is recognized based on the measurement result of the distance measuring sensor 212a and the camera image, only different portions will be described.
FIG. 42 is a diagram illustrating the processing result of the camera image.

  In the sensing processor 210, first, an image is acquired from the camera 222 through step S250. For example, assume that an image as shown in FIG. Next, through step S252, distortion correction processing is executed on the camera image, and an image with corrected distortion is obtained as shown in FIG. Next, through step S254, edge extraction processing is performed on the camera image whose distortion has been corrected, and an image in which only edges are extracted is obtained as shown in FIG. In step S256, a line segment is detected from the edge-extracted camera image. As a result, six line segments are obtained as shown in FIG.

Next, through steps S300 to S326, measurement results by the first scanning process are obtained for all scanning planes.
When the measurement is completed for all the scanning planes, the sensor feature points are projected onto the camera image through coordinate transformation based on the coordinate information stored in the memory through step S328.
FIG. 43 is a diagram illustrating the determination result of the continuous surface.

FIG. 44 is a diagram illustrating a result of projecting sensor feature points onto a camera image.
For the scanning plane φ0, as shown in FIG. 43, (1) the intersection of the region (φ0, 0) corresponding to the floor surface and the region (φ0, 1) corresponding to the first stage kick plate, (2) The intersection of the region (φ0, 1) and the region (φ0, 2) corresponding to the first step plate, (3) the region (φ0, 2) and the region (φ0, 3) corresponding to the second step kick plate And four intersections of (4) the intersection of the region (φ0, 3) and the region (φ0, 4) corresponding to the second step board are determined as sensor feature points. Therefore, as shown in FIG. 44, these four sensor feature points are projected onto the camera image as projection points on the scanning plane φ0 that extends from the upper right corner to the lower left corner.

  For the scanning plane φ1, as shown in FIG. 43, (1) the intersection of the region (φ1, 0) corresponding to the floor surface and the region (φ1, 1) corresponding to the first-stage kick plate, (2) The intersection of the region (φ1, 1) and the region (φ1, 2) corresponding to the first step board, (3) the region (φ1, 2) and the region (φ1, 3) corresponding to the second step kick plate And four intersections of (4) the intersection of the region (φ1, 3) and the region (φ1, 4) corresponding to the second step board are determined as sensor feature points. Therefore, as shown in FIG. 44, these four sensor feature points are projected onto the camera image as projection points on the scanning plane φ1 that extends from the position directly above the center to the position directly below the center.

  For the scanning plane φ2, as shown in FIG. 43, (1) the intersection of the region (φ2, 0) corresponding to the floor surface and the region (φ2, 1) corresponding to the first stage kick plate, (2) The intersection of the region (φ2, 1) and the region (φ2, 2) corresponding to the first step plate, (3) The region (φ2, 2) and the region (φ2, 3) corresponding to the second step kick plate And four intersections of (4) the intersection of the area (φ2, 3) and the area (φ2, 4) corresponding to the second step board are determined as sensor feature points. Therefore, as shown in FIG. 44, these four sensor feature points are projected onto the camera image as projection points on the scanning plane φ2 that extends from the upper left center to the lower right center.

  Next, through steps S330 to S334, in the camera image, the line segment of the camera image is associated with the sensor feature point, the line segments associated with the same sensor feature point are grouped, and the line segments belonging to the same group are grouped. The boundary line of the continuous surface is determined by linear approximation based on the coordinates of the end points. In the example of FIG. 44, two straight lines are determined as the boundary line of the continuous surface from the positional relationship between the line segment of the camera image and each projection point.

Then, through steps S337 and S339, boundary data is generated based on the information about the straight line determined as the boundary line of the continuous surface, and the generated boundary data is output to the CPU 60.
When the boundary data is input, the CPU 60 detects the feature points of the stairs based on the input boundary data through step S402. Further, the input boundary data is analyzed, and for example, a continuous surface having an inclination close to 0 is regarded as a horizontal surface, and it is determined that the leg-wheel type robot 100 is a surface on which walking is possible.

  Thus, in this embodiment, the distance measuring sensor 212a that measures the distance to the measurement point on the object, and the camera 222 that captures an image including the measurement point that can be measured in the scanning range of the distance measuring sensor 212a. The distance measurement sensor 212a is scanned, the measurement result of the distance measurement sensor 212a is obtained for the measurement points that can be measured in the scanning range, the sensor feature point is detected based on the obtained measurement result, and the image is taken from the camera 222. , The line segment is detected from the camera image, and the boundary of the continuous surface is recognized based on the line segment of the detected camera image and the sensor feature point.

  Thereby, in addition to the distance measurement sensor 212a, a feature on the object is detected using a camera 222 of a different system, and the boundary of the continuous surface is recognized based on the two detection results. The shortcomings of the system based on 212a can be compensated for by the system based on the camera 222, and the possibility of a decrease in recognition accuracy can be reduced when object recognition is performed by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a. . In addition, since it is not necessary to increase the number of scans in the second scanning process, the measurement time can be shortened.

Further, in the present embodiment, the distance measurement sensor 212a is scanned, the measurement result of the distance measurement sensor 212a is obtained for the measurement points that can be measured in the scanning range, and the obtained measurement result is converted into the coordinates of the orthogonal coordinate system. The line segment in the orthogonal coordinate system is detected by the Hough transform based on the converted coordinates of at least two measurement points, and the sensor feature point is detected based on the detected line segment.
This makes it possible to grasp at least the planar shape of an object as a boundary between continuous surfaces, which is suitable for posture control of robots that require complex posture control, such as legged robots and leg-wheel type robots 100. A recognition result can be obtained. Further, since the line segment is detected by the Hough transform, when the object recognition is performed by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a, there is a possibility that the recognition accuracy may be lowered due to a decrease in measurement resolution or a variation in measurement results. Can be reduced.

Furthermore, as compared with the least square method, robust detection can be performed for an edge that is difficult to detect.
Furthermore, in this embodiment, sensor feature points are detected based on the coordinates of the end points of the detected line segment.
Thereby, the boundary of a continuous surface can be recognized comparatively correctly.

Furthermore, in this embodiment, a line segment is detected from the camera image, and the boundary of the continuous surface is recognized in the camera image based on the positional relationship between the line segment of the camera image and the sensor feature point.
Thereby, possibility that recognition accuracy will fall can further be reduced.
Furthermore, in this embodiment, for each line segment of the camera image, the line segment is associated with a sensor feature point existing within a predetermined distance from a straight line obtained by extending the line segment, and the same sensor feature point corresponds The attached line segments are grouped, and the boundary line of the continuous surface is determined based on the coordinates of the end points of the line segments belonging to the same group.

As a result, due to the influence of contrast or the like, even though a plurality of line segments should be detected from the camera image as actually one line segment, a plurality of line segments are set to 1 based on the correspondence with sensor feature points. Since it is grouped as one line segment and the boundary line of the continuous surface is determined, the possibility that the recognition accuracy is lowered can be further reduced.
In the fifth embodiment, the camera 222 corresponds to the imaging means of the invention 15, 22 or 24, and the yrs' axis rotation mechanism is the first scanning means of the invention 19 or 20, or the first rotation of the invention 20. The zrs axis rotation mechanism corresponds to the second scanning means of the invention 19 or 20, or the second rotation means of the invention 20. Step S250 corresponds to the imaging step of the invention 27, steps S252 to S256 correspond to the second feature detecting means of the invention 15, 16 or 22, or the second feature detecting step of the invention 27, and step S256 is This corresponds to the image line segment detection means of the invention 22 or 23.

  In the fifth embodiment, steps S306, S310, and S320 correspond to the first feature detecting means of the invention 15, 16, or 22, or the first feature detecting step of the invention 27, and steps S302 and S340 are This corresponds to the measurement result acquisition means of the invention 15, 16, 19, 20 or 22, or the measurement result acquisition step of the invention 27. Step S306 corresponds to the coordinate transformation means of the invention 16, step S310 corresponds to the line segment detection means of the invention 16, and step S334 is the recognition means of the invention 15, 16, 22 or 23, or the invention 27. This corresponds to the recognition step.

In the fifth embodiment, the yrs ′ axis corresponds to the first scanning axis of the invention 20 and the zrs axis corresponds to the second scanning axis of the invention 20.
[Sixth Embodiment]
Next, a sixth embodiment of the present invention will be described with reference to the drawings. FIG. 45 is a diagram showing a sixth embodiment of the leg-wheel robot, object recognition device, and object recognition method according to the present invention.

This embodiment is different from the fifth embodiment in that the Hough transform is performed after connecting the measurement points with line segments in the orthogonal coordinate system. Hereinafter, only the parts different from the fifth embodiment will be described, and the same parts as those in the fifth embodiment will be denoted by the same reference numerals and the description thereof will be omitted.
First, an object recognition process executed by the three-dimensional distance measuring apparatus 200 will be described.

FIG. 45 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200.
When the object recognition process is executed in the sensing processor 210, as shown in FIG. 45, the process proceeds to step S308 through steps S250 to S306.
In step S308, based on the coordinate information converted in step S306, measurement points adjacent in the x-axis direction in the scanning plane coordinate system are connected with line segments, and the process proceeds to step S310 and the obtained line is changed. A line segment in the orthogonal coordinate system is detected by Hough transform based on the coordinate information of the point.

Then, after steps S320 and S324, the process proceeds to step S326, where it is determined whether or not the processing of steps S302 to S324 has been completed for all the scanning planes, and when it is determined that the processing has been completed (Yes) Through steps S328 to S339, the series of processes is terminated.
On the other hand, when it is determined in step S326 that the processes in steps S302 to S324 are not completed for all scanning planes (No), the process proceeds to step S302 through step S340.

Next, the operation of the present embodiment will be described.
A case will be described in which a stairway exists on the movement path of the leg-wheel type robot 100 and the stairs are overcome. Since the fifth embodiment is different from the fifth embodiment in that a line segment is detected by Hough transform, only different portions will be described.
Through steps S308 and S310, measurement points are connected by line segments based on the converted coordinate information, and line segments in the orthogonal coordinate system are detected by Hough transform based on the coordinate information of the points on the obtained line. .

  In the orthogonal coordinate system, each measurement point is represented as a set of a plurality of points along the contours of the kick board and the tread board as shown in FIG. In the example of FIG. 36 (a), in the region corresponding to the first step tread, some measurement points are distributed in a region deviating from the continuous surface. The measurement results vary due to the influence of the error area A1. The area A2 has a lower measurement resolution than the other areas.

When a line segment connection is made for this measurement result, each measurement point is connected by one line as shown in FIG. As a result, between the measurement points where no measurement point exists, the points on the line are interpolated.
When the Hough transform is performed on the connection result, as shown in FIG. 36 (c), a line segment along the outline of each kick board and each tread board is detected.

If the Hough transform is performed on the measurement result in FIG. 36A without connecting the line segments, as shown in FIG. 33B, line segments along the contours of the treads are detected. The line segment along each kick plate cannot be detected.
As described above, in this embodiment, the distance measuring sensor 212a that measures the distance to the measurement point on the object is provided, the distance measurement sensor 212a is scanned, and the distance measurement sensor can be measured for the measurement points that can be measured in the scanning range. The measurement result of 212a is acquired, the acquired measurement result is converted into coordinates in an orthogonal coordinate system, the converted measurement points are connected by line segments, and orthogonality is obtained by Hough transform based on the coordinates of the points on the obtained line A line segment in the coordinate system is detected, and the boundary of the continuous surface is recognized based on the detected line segment.

  This makes it possible to grasp at least the planar shape of an object as a boundary between continuous surfaces, which is suitable for posture control of robots that require complex posture control, such as legged robots and leg-wheel type robots 100. A recognition result can be obtained. Further, since the line segment is detected by the Hough transform, when the object recognition is performed by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a, there is a possibility that the recognition accuracy may be lowered due to a decrease in measurement resolution or a variation in measurement results. Can be reduced. In addition, since line segments are detected based on the coordinates of points on the line obtained by connecting the measurement points with line segments, relatively accurate recognition results can be obtained even if measurement resolution decreases or measurement results vary. The possibility that the recognition accuracy is lowered can be further reduced.

Furthermore, as compared with the least square method, robust detection can be performed for an edge that is difficult to detect.
In the sixth embodiment, steps S252 to S256 correspond to the second feature detection means of the invention 17, steps S306 to S310, S320 correspond to the first feature detection means of the invention 17, and steps S302, S340 corresponds to the measurement result acquisition means of the seventeenth aspect. Step S306 corresponds to the coordinate transformation means of the invention 17, step S308 corresponds to the inter-measurement point interpolation means of the invention 17, step S310 corresponds to the line segment detection means of the invention 17, and step S334 is This corresponds to the recognition means of the invention 17.
[Seventh Embodiment]
Next, a seventh embodiment of the present invention will be described with reference to the drawings. 46 to 49 are diagrams showing a seventh embodiment of the leg-wheel type robot, the object recognition apparatus, and the object recognition method according to the present invention.

  Compared to the fifth embodiment, the present embodiment calculates the inclination at each measurement point in the orthogonal coordinate system, calculates the appearance frequency of each inclination, the calculated appearance frequency and the coordinates of each measurement point The point of recognizing the boundary of the continuous surface is different. Hereinafter, only the parts different from the fifth embodiment will be described, and the same parts as those in the fifth embodiment will be denoted by the same reference numerals and the description thereof will be omitted.

First, an object recognition process executed by the three-dimensional distance measuring apparatus 200 will be described.
FIG. 46 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200.
When the object recognition process is executed in the sensing processor 210, the process proceeds to step S508 through steps S250 to S306 as shown in FIG.

In step S508, the inclination at each measurement point in the orthogonal coordinate system is calculated based on the coordinate information converted in step S306.
FIG. 47 is a diagram illustrating an example of calculating the inclination. In the figure, the x-axis and z-axis indicate axes in the scanning plane coordinate system, and are separate from the axes in the sensor coordinate system and the camera coordinate system.
The inclination at the measurement point is shown in FIG. 47 (a) based on the coordinates of the measurement point and the coordinates of a predetermined number of measurement points adjacent in the x-axis direction for each measurement point in the scanning plane coordinate system. As shown, it can be calculated by the least square method.

As shown in FIG. 47 (a), a regression line is approximated to a predetermined number of measurement points before and after the measurement points (white circles in FIG. 47 (a)) to be calculated by the least square method. The inclination is the inclination at the target measurement point. The least squares method is less affected by data variations but has a relatively large amount of calculation.
As another method, for each measurement point, an inclination can be calculated based on a difference value between coordinates of two measurement points before and after a predetermined number of other measurement points.

As shown in FIG. 47 (b), a first difference value that is a difference value between z coordinates of two measurement points before and after (for example, if the z coordinate of the measurement point of interest is z _i , z _{i + 2} −z _{i -2)} calculates the second difference value is a difference value between the x-coordinate of the front and rear two measurement points (e.g., the x-coordinate of the measurement point of interest When _{x i, x i + 2 -x} i-2 ) And the inclination (first difference value / second difference value) at the measurement point of interest is obtained from these calculation results. In this method, the influence of data variation is greater than that in the least squares method, but since the calculation requires only simple subtraction and division, the slope can be calculated faster than in the least squares method.

Next, as shown in FIG. 46, the process proceeds to step S510, and the appearance frequency of each inclination with respect to the total number of inclinations calculated in step S508 is calculated. For example, the number of measurement points at each inclination with respect to the total number of measurement points is calculated as the appearance frequency.
Next, the process proceeds to step S512, where the appearance frequency of the unprocessed inclination is selected from the appearance frequencies of the respective slopes calculated in step S510, and the process proceeds to step S514, where the selected appearance frequency and a preset threshold value are selected. Are determined, and it is determined whether or not the selected appearance frequency is equal to or higher than the threshold value.

In step S516, based on the coordinate information of the measurement points having the inclination related to the selected appearance frequency and the inclination of the predetermined range, it is determined that the area including the measurement points having continuous coordinates is a continuous surface. In a region determined to be a continuous surface, a measurement point whose inclination changes abruptly is determined as a boundary of the continuous surface.
Next, the process proceeds to step S518, where the coordinate information of the measurement point determined as the boundary of the continuous surface is converted into the sensor coordinate system, the converted coordinate information is stored in the memory, and the process proceeds to step S520.

In step S520, it is determined whether or not the processing in steps S514 to S518 and S534 has been completed for all inclinations. If it is determined that the processing has been completed (Yes), the process proceeds to step S326.
In step S326, it is determined whether or not the processing in steps S302 to S520 has been completed for all the scanning planes. When it is determined that the processing has been completed (Yes), the series of processing is completed through steps S328 to S339. To do.

On the other hand, when it is determined in step S326 that the processes in steps S302 to S520 are not completed for all scanning planes (No), the process proceeds to step S302 via step S340.
On the other hand, when it is determined in step S520 that the processes in steps S514 to S518 and S534 are not completed for all inclinations (No), the process proceeds to step S512.

On the other hand, when it is determined in step S514 that the selected appearance frequency is less than the threshold value (No), the process proceeds to step S534, and the coordinate information of the measurement point having the inclination related to the selected appearance frequency is excluded as measurement noise. Then, the process proceeds to step S520.
Next, the operation of the present embodiment will be described.
A case will be described in which a stairway exists on the movement path of the leg-wheel type robot 100 and the stairs are overcome. Since the fifth embodiment is different from the fifth embodiment in that the boundary of the continuous surface is determined based on the appearance frequency of the inclination, only different portions will be described.

First, through step S508, the inclination at each measurement point is calculated based on the converted coordinate information.
FIG. 48 is a diagram illustrating an example of inclinations in the x-axis direction and the z-axis direction calculated by the least square method.
Here, the slope at each measurement point is calculated by the least square method. Here, when the least square method is used, as shown in FIGS. 48A and 48B, the inclination in the x-axis direction and the z-axis direction with respect to a plurality of measurement points to which the least square method is applied. May be different from the slope. Therefore, as shown in the following equations (19) and (20), the least square method calculation is performed using two linear equations for obtaining the inclinations in the x-axis direction and the z-axis direction.

z = a _x · x + b _x (19)
x = _az , z + _bz (20)

Then, a _x and b _x in the above equation (19) are calculated by the following equation (21), and a _z and b _z in the above equation (20) are calculated by the following equation (22).

When a _x and b _x , and a _z and b _z are calculated, the distance between the straight line of the above equations (19) and (20) and the respective measurement points used for the least square calculation is calculated by substituting these calculation results. While calculating as residuals h _x and h _z , the sum of squares of the residuals h _{x and} the sum of squares of the residuals h _z for the respective measurement points are calculated (the following equation (23)).

When the square sum of the residual h _{x and} the square sum of the residual h _z are calculated, the straight line with the smaller value is compared as the correct straight line, and the calculation target is calculated based on the slope of this straight line. Find the slope at the measurement point. Specifically, _if the square sum of the residual h _x is smaller, “a = a _x , b = b _x ”. _If the square sum of the residual h _z is smaller, “a = 1 / _az”. , B = 1 / b _z ”.

  Next, through steps S510 to S514, the appearance frequency of each inclination is calculated, the appearance frequency of the undetermined inclination is selected, and it is determined whether or not the selected appearance frequency is equal to or greater than the threshold value. At this time, if it is equal to or greater than the threshold value, the boundary of the continuous surface is determined through step S516 based on the selected inclination of the appearance frequency and the coordinate information of the measurement point having the predetermined range of inclination.

FIG. 49 is a diagram illustrating an example of the appearance frequency of an inclination with respect to a certain scanning plane.
For example, it is assumed that the appearance frequency as shown in FIG. 49 is obtained. In the example of FIG. 49, the appearance frequency near the inclination of 0 is relatively high, and if the coordinates of the measurement points having these inclinations are continuous, the region including the continuous measurement points is a continuous surface. It is determined. On the movement path of the leg-wheel type robot 100, there is a staircase having a constant step and a tread plate as a continuous surface. Therefore, the horizontal plane corresponding to the tread plate has a series of measurement points with zero inclination and substantially zero. Appear as coordinates. In this case, it can be estimated that there is a horizontal plane there.

  And in the area | region determined to be a continuous surface, the measurement point in which inclination changes rapidly is determined as a boundary of a continuous surface. In the example of FIG. 43, for a certain scanning plane, (1) the intersection of the area corresponding to the floor surface and the area corresponding to the first-stage kick board, (2) the area and the area corresponding to the first-stage tread board, The four intersections of the intersection of (3) the area and the area corresponding to the second-stage kick board, and (4) the intersection of the area and the area corresponding to the second-stage tread are Determined.

Next, through step S518, the coordinate information of the measurement point determined as the boundary of the continuous surface is converted into the sensor coordinate system, and the converted coordinate information is stored in the memory.
On the other hand, if it is less than the threshold value, the coordinate information of the measurement point is excluded as measurement noise through step S534.
The processes in steps S514 to S518 and S534 are performed for all inclinations.

  As described above, in this embodiment, the distance measuring sensor 212a that measures the distance to the measurement point on the object is provided, the distance measurement sensor 212a is scanned, and the distance measurement sensor can be measured for the measurement points that can be measured in the scanning range. The measurement result of 212a is acquired, the acquired measurement result is converted into the coordinates of the orthogonal coordinate system, and for each converted measurement point, the orthogonal coordinate system is based on the coordinates of the measurement point and the coordinates of the surrounding measurement points. The slope at the measurement point is calculated, the appearance frequency of each slope with respect to the calculated total number of slopes is calculated, and the boundary of the continuous surface is recognized based on the calculated appearance frequency and the converted coordinates of each measurement point.

  This makes it possible to grasp at least the planar shape of an object as a boundary between continuous surfaces, which is suitable for posture control of robots that require complex posture control, such as legged robots and leg-wheel type robots 100. A recognition result can be obtained. Further, since the boundary of the continuous surface is recognized based on the appearance frequency of the inclination at the measurement point and the coordinates of each measurement point, the recognition is performed when the object is recognized by the two-dimensional distance measuring device 212 using the distance measuring sensor 212a. The possibility that the accuracy is lowered can be reduced.

  In the seventh embodiment, steps S252 to S256 correspond to the second feature detection means of the invention 18, steps S306 and S508 to S516 correspond to the first feature detection means of the invention 18, and steps S302, S340 corresponds to the measurement result acquisition means of the eighteenth aspect. Step S306 corresponds to the coordinate conversion means of the invention 18, step S508 corresponds to the inclination calculation means of the invention 18, step S510 corresponds to the appearance frequency calculation means of the invention 18, and step S334 corresponds to the invention. This corresponds to 18 recognition means.

In the third and fourth embodiments, the configuration is such that the continuous surface on the object or the boundary of the continuous surface is recognized. However, the present invention is not limited to this. It can also be configured to recognize.
In the fourth and sixth embodiments, adjacent measurement points are connected by line segments. However, the present invention is not limited to this, and the measurement points may be measured by straight lines, multi-degree curves or other curves. May be connected, or the measurement points may be approximated by a straight line, a line segment, a multi-order curve, or other curves. In this case, the measurement points do not necessarily have to be located on the line, and adjacent measurement points need not be targeted.

In the fourth and sixth embodiments, the line segment is detected by the Hough transform based on the coordinates of the points on the line obtained by the line segment connection. However, the present invention is not limited to this. A line segment may be detected by Hough transform based on the coordinates of a point existing within a predetermined distance from the given line.
In the fifth to seventh embodiments, the three-dimensional distance measuring device 200 rotates the distance measuring sensor 212a around the yrs' axis and rotates the two-dimensional distance measuring device 212 around the zrs axis. However, the present invention is not limited to this, and includes a plurality of distance measuring sensors 212a and a plurality of rotation mechanisms that respectively rotate the distance measuring sensors 212a around a scanning axis orthogonal to the measurement direction of the distance measuring sensor 212a. The distance measuring sensor 212a may be arranged around the zrs axis.

As a result, as in the fifth to seventh embodiments, the three-dimensional shape of the object can be grasped as the boundary of the continuous surface, so that it is complicated as in the legged robot and the leg wheel type robot 100. A recognition result suitable for posture control of a robot that requires posture control can be obtained.
In the fifth to seventh embodiments, the line segment is detected from the camera image by the Hough transform. However, the present invention is not limited to this, and the line segment is detected by the following detection method. It can also be configured.

As a first detection method, the edge extraction process is executed a plurality of times while changing the parameters of the edge extraction process, and a line segment is detected by Hough transform from each edge-extracted camera image. And the boundary of a continuous surface is recognized based on all the detected line segments. Thereby, possibility that recognition accuracy will fall can further be reduced.
As a second detection method, a plurality of images are acquired from the camera 222 by changing the shooting conditions of the camera 222 such as a shutter speed and an aperture, and distortion correction processing and edge extraction processing are respectively performed on the camera images. A line segment is detected by Hough transform from each camera image obtained by edge extraction. And the boundary of a continuous surface is recognized based on all the detected line segments. Thereby, possibility that recognition accuracy will fall can further be reduced.

In the fifth to seventh embodiments, the boundary of the continuous surface on the object is recognized. However, the present invention is not limited to this, and the continuous surface, intermittent surface, and other surfaces on the object, or their boundaries. It can also be configured to recognize.
In the fifth to seventh embodiments, the boundary data is generated on the basis of the information about the straight line determined as the boundary line of the continuous surface. It can also be configured to generate the surface data based on the information regarding the straight line determined as the boundary line and the coordinate information of the sensor feature point.

  Since the line segment connecting the sensor feature points determined as the boundary of the continuous surface is an intersection line where the continuous surface and the scanning plane intersect, generation of the surface data includes, for example, a line segment connecting the sensor feature points in a certain scanning plane, In the scanning plane adjacent to the z-axis, those having inclinations and coordinates with a line segment connecting sensor feature points within a predetermined range are determined as continuous surfaces, and correspond to the straight line determined as the boundary line between these line segments and the continuous surface. This is performed by associating coordinate information or connecting them using a known interpolation method. For example, since a continuous surface having an inclination close to 0 can be regarded as a horizontal surface, it can be determined that the surface is a walking surface.

  In the fifth to seventh embodiments, the sensor feature points are projected onto the camera image by coordinate transformation. However, the present invention is not limited to this, and the line segment of the camera image is represented by a plane in the scanning plane coordinate system. The sensor feature point and the line segment of the camera image may be projected onto a plane or space of another coordinate system. That is, the coordinate system for determining the positional relationship between the line segment of the camera image and the sensor feature point can be arbitrary.

  In the fifth to seventh embodiments, the line segment is detected from the camera image and the boundary of the continuous surface is recognized based on the detected line segment. However, the present invention is not limited to this. Detect points, straight lines, multi-dimensional curves, circles, ellipses and other curves, planes, curved surfaces and other surfaces, cubes, spheres and other three-dimensional features from the camera image, and continuous surfaces based on the detected features It is also possible to configure so as to recognize the boundary.

  In the fifth to seventh embodiments, the sensor feature point is detected based on the measurement result of the distance measuring sensor 212a, and the boundary of the continuous surface is recognized based on the detected sensor feature point. However, the present invention is not limited to this, and straight lines, line segments, multi-order curves, circles, ellipses and other curves, planes, curved surfaces and other surfaces, cubes, spheres, and other three-dimensional features that indicate features on the object are detected and detected. The boundary of the continuous surface can be recognized based on the feature.

  In the fifth to seventh embodiments, the camera 222 is arranged so that the xrs-yrs plane in the sensor coordinate system and the xca-yca plane in the camera coordinate system are parallel to each other. However, the present invention is not limited to this. First, if an image including a measurement point that can be measured within the scanning range of the distance measuring sensor 212a can be taken, the image can be arranged at an arbitrary position and orientation with respect to the distance measuring sensor 212a.

In the third to seventh embodiments, the encoder 218 is provided on the drive rotary shaft. However, the present invention is not limited to this, and the encoder 218 may be provided on the driven rotary shaft.
Thereby, the scanning angle in the second scanning process can be detected with high accuracy without being affected by the transmission error.
In the third to seventh embodiments, the rotational driving force of the drive rotating shaft is transmitted to the driven rotating shaft via the pulleys 220a and 220b and the belt 221. Not limited to this, the configuration may be such that the rotational driving force of the drive rotary shaft is transmitted to the driven rotary shaft via a plurality of gears.

  In the third to seventh embodiments, the distance information is acquired in the first scanning process discretely (distance information is acquired in the first scanning process after being rotated in the second scanning process). However, the present invention is not limited to this, and the configuration is such that the scanning angle and the measurement distance are associated with each other continuously (distance information is acquired by the first scanning process while being rotated by the second scanning process). You can also.

In the third to seventh embodiments, the scanning angle range and the scanning unit angle are set based on the command signal from the CPU 60. However, the present invention is not limited to this, and the three-dimensional distance measurement is performed. A configuration in which the apparatus 200 is set in advance may be employed.
In the third to seventh embodiments, the three-dimensional distance measuring device 200 rotates the distance measuring sensor 212a around the yrs' axis and rotates the two-dimensional distance measuring device 212 around the zrs axis. However, the present invention is not limited to this, and the distance measuring sensor 212a may be rotated about the zrs axis, and the two-dimensional distance measuring device 212 may be rotated about the yrs' axis. Further, not only about the yrs ′ axis and the zrs axis, but any two axes that are orthogonal to the measurement direction of the distance measuring sensor 212a may be used. Further, not limited to such a rotation mechanism, the distance measuring sensor 212a is moved in a first scanning direction that is different from the measuring direction of the distance measuring sensor 212a, and the measuring direction of the distance measuring sensor 212a is different from the first scanning direction. It can also be configured as a moving mechanism so as to move the distance measuring sensor 212a in two scanning directions. In this case, as the shape of the movement path, a circular arc or other curve can be adopted in addition to a straight line. A combination of a rotating mechanism and a moving mechanism can also be used.

Such a configuration change can be similarly applied to the above-described configuration including the plurality of distance measuring sensors 212a and the rotation mechanism.
In the fifth to seventh embodiments, as shown in FIG. 50A, the two-dimensional distance measuring device 212 itself is rotated. However, the present invention is not limited to this, and an optical distance measuring sensor is used. As shown in FIG. 50B, a mirror inserted on the optical axis in the measurement direction may be rotated.

FIG. 50 is a diagram illustrating a configuration when the measurement direction of the distance measuring sensor is changed.
Moreover, although the said 3rd thru | or 7th embodiment was demonstrated as an independent embodiment, the said 3rd thru | or 7th embodiment is applied with respect to the said 1st or 2nd embodiment. You can also
In the third to seventh embodiments, the leg-wheel type robot, the object recognition apparatus, and the object recognition method according to the present invention are applied to the case of going over stairs. The same applies to the case of overcoming the step.

  In the third to seventh embodiments, the leg-wheel type robot, the object recognition apparatus, and the object recognition method according to the present invention are applied to the leg-wheel type robot 100. However, the present invention is not limited to this. The present invention can be applied to other cases without departing from the spirit of the invention. For example, the present invention can be applied to a leg-wheel type robot that realizes a leg structure with a linear motion joint, a robot having another configuration, a vehicle, and other devices. Further, by combining with a means for transmitting recognition information, development to a device and a recognition method as a means for a visually impaired person to recognize the outside world is also conceivable.

1 is a front view of a leg wheel type robot 100. FIG. 1 is a side view of a leg wheel type robot 100. FIG. It is a figure which shows the structure of the obstruction sensors 34 and 36. FIG. 2 is a block diagram showing a movement control system of a leg wheel type robot 100. FIG. FIG. 3 is a diagram showing a running state of a leg wheel type robot 100. It is a flowchart which shows a raising / lowering control process. It is a figure for demonstrating the principle of a light cutting method. It is a figure which shows the state which irradiated the laser beam to the staircase, and the image of the image pick-up element of the camera. It is a figure which shows operation | movement in case each drive wheel 20 turns without grounding on a yaw axis. It is a figure which shows operation | movement in case each drive wheel 20 grounds on a yaw axis and turns. It is a figure which shows the operation | movement in an area with a low ceiling where vehicle height is restrict | limited. (A)-(c) is a figure which shows the example of a driving | running | working state of the leg wheel type robot 100 at the time of non-turning driving | running | working control. (A) And (b) is a figure which shows the driving | running | working state of the leg wheel type robot 100 at the time of super-trust turning control in case the coordinate (0, 0) and coordinate (xc, yc) of the base | substrate 10 are made into a rotation center. It is. (A) And (b) is a figure which shows an example of the driving | running route of the leg wheel type robot 100. FIG. It is a figure which shows an example of the gravity center position of a robot. 1 is a front view of a leg wheel type robot 100. FIG. 1 is a side view of a leg wheel type robot 100. FIG. It is a front view (yrs-zrs plane) of the three-dimensional distance measuring apparatus 200. It is a side view (xrs-zrs plane) of the three-dimensional distance measuring apparatus 200. It is a top view (yrs-xrs plane) of the three-dimensional distance measuring apparatus 200. It is a figure which shows the scanning range of a ranging sensor. 2 is a block diagram showing a movement control system of a leg wheel type robot 100. FIG. It is a block diagram which shows the control structure of the two-dimensional distance measuring apparatus 212. 3 is a block diagram showing a control structure of a three-dimensional distance measuring apparatus 200. FIG. It is a figure for demonstrating the principle of the distance measurement of the two-dimensional distance measuring apparatus 212. FIG. It is a figure which shows the case where it scans by the 1st scanning process and the 2nd scanning process. It is a figure which shows the example of a distance measurement of the three-dimensional distance measuring apparatus 200. FIG. 4 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200. It is a figure for demonstrating the principle of Hough conversion. It is a flowchart which shows a raising / lowering control process. It is a graph which shows the measurement result by the 1st scanning process. It is a graph which shows the result of having converted the distance information of the rotation coordinate system of each measurement point of FIG. 31 into the coordinate information of a rectangular coordinate system. It is a graph which shows the measurement point in a rectangular coordinate system, and the result of Hough transform. It is a figure which shows the determination result of a continuous surface. 4 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200. It is a graph which shows the measurement point in a rectangular coordinate system, the result of line segment connection, and the result of Hough transform. It is a figure for demonstrating the principle of the object recognition using the camera 222 and the three-dimensional distance measuring apparatus 200. FIG. 4 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200. 2 is a diagram illustrating a sensor coordinate system, a camera coordinate system, and a coordinate system of an image sensor surface of a camera 222. FIG. It is a figure for demonstrating the case where the matching of the segment of a camera image and a sensor feature point, grouping, and the determination of a boundary line are performed. It is a flowchart which shows a raising / lowering control process. It is a figure which shows the processing result of a camera image. It is a figure which shows the determination result of a continuous surface. It is a figure which shows the projection result to the camera image of a sensor feature point. 4 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200. 4 is a flowchart showing object recognition processing executed by the three-dimensional distance measuring apparatus 200. It is a figure which shows the example of calculation of inclination. It is a figure which shows an example of the inclination of the x-axis direction computed by the least square method, and az-axis direction. It is a figure which shows an example of the appearance frequency of the inclination with respect to a certain scanning plane. It is a figure which shows the structure in the case of changing the measurement direction of a ranging sensor.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Leg wheel type robot 10 Base 12 Leg part 14-18 Rotary joint 20 Driving wheel 22, 24 Leg tip sensor 26 Horizontal laser 28, 30 Vertical laser 34, 36 Obstacle sensor 40, 50 Motor 42, 52 Encoder 44, 54 Driver 70 Three-axis attitude sensor 200 Three-dimensional distance measurement device 204 Lower support plate 206 Upper support plate 208 Camera support plate 210 Sensing processor 212 Two-dimensional distance measurement device 212a Distance sensor 212c, 216 Motor 212d, 218 Encoder 212b, 214 Driver 220a 220b Pulley 221 Belt 32, 222 Camera

Claims (8)

A base, a leg connected to the base with at least a degree of freedom around the yaw axis, a wheel rotatably provided on the leg, and a range of freedom around the yaw axis A leg-wheel type robot that includes an actuator for applying power for driving the leg and a control means for controlling the actuator, and moves by driving the leg and rotating the wheel;
The leg wheel type robot characterized in that the control means performs drive control of the leg portion at the time of obstacle avoidance and steering control of the wheel at the time of running by using the actuator in common. A base, a leg connected to the base with a degree of freedom around the yaw axis and a degree of freedom around the pitch axis or roll axis, a wheel rotatably provided on the leg, and the yaw A first actuator that applies power for driving the leg in a range of degrees of freedom around the axis, and power for driving the leg in a range of degrees of freedom around the pitch axis or roll axis A leg wheel type robot comprising a second actuator, and a control means for controlling the first actuator and the second actuator, wherein the leg wheel type robot moves by driving the leg and rotating the wheel;
The control means uses the first actuator and the second actuator to perform drive control of the leg when avoiding an obstacle, and performs steering control of the wheel during travel using the first actuator. A leg-wheel type robot characterized by performing. In claim 2,
A plurality of the leg portions, and a plurality of the wheels provided rotatably on each leg portion,
The control means controls the first actuator and the second actuator so that a positional relationship between a polygon formed by a ground contact point of each wheel and a projected point of the center of gravity of the base body is constant during turning. A leg-wheel robot characterized by In claim 3,
The control means controls the first actuator and the second actuator so that a polygonal center formed by a ground contact point of each wheel coincides with a projected point of the center of gravity of the base body at the time of turning. A leg-wheel type robot characterized by In any one of Claims 3 and 4,
The control means includes the first actuator and the second actuator so that, when turning, the wheel is positioned on the shaft of a portion having a degree of freedom around the yaw axis in a leg portion provided with the wheel. A leg-wheel type robot characterized by controlling an actuator. In any one of Claims 2 thru | or 5,
The control means, when displacing the base body in the vertical direction, is configured so that the positional relationship between the polygon formed by the ground contact points of the wheels and the projected point of the center of gravity of the base body is constant. A leg-wheel type robot that controls an actuator and the second actuator. In any one of Claims 2 thru | or 6,
The control means controls the first actuator so that the traveling direction of the self-legged wheel type robot and the traveling direction of each wheel coincide with each other while maintaining the orientation of the base body in a constant direction. Leg-wheel type robot. In any one of Claims 2 thru | or 7,
A plurality of the leg portions, the wheels, the leg wheel mechanism portion composed of the first actuator and the second actuator with respect to the base body,
The control means includes a contact point between an arc trajectory drawn by a rotation center at the time of steering of each wheel and a rotation center at the time of steering when the base body rotates around the yaw axis at a predetermined rotation center position during turning. A leg-wheel robot, wherein the first actuator is controlled so that a movement direction of the rotation center at a position coincides with a traveling direction of each wheel.

Images