JP2017102705A - Autonomous mobile device and autonomous mobile device system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an autonomous mobile device and an autonomous mobile device system that can improve estimation precision of the position of the autonomous mobile device.SOLUTION: The present invention relates to a self-traveling robot 1 as an autonomous mobile device comprising: a driving motor and driving wheels 3 as moving means; a map storage part as map storage means of storing map information on a movement region; a region measuring sensor 5 as range finding means of measuring distances to two kinds of a two-stage shelf 40 and a three-stage shelf 50 present in the movement region; and a computation processing part 10 as estimation means of estimating a self-position in the movement region by collating contour information on circumferential shelves found from a measurement result of the region measuring sensor 5 with contour information on shelves in the movement region included in map information. The region measuring sensor 5 can be changed in height between H1 and H2, and the map storage part comprises, as map information, two-dimensional plane map information of height H1 in a first region 400 and two-dimensional plane map information of height H2 in a second region 500.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an autonomous mobile device and an autonomous mobile device system.

Conventionally, as an autonomous mobile device, a distance sensor detects the azimuth and distance to a surrounding object, and based on the detection result and the position information of the object in the map information of the movement area stored in advance, the self-position It is known to estimate
For example, Patent Document 1 describes the following autonomous traveling device.
The laser distance sensor detects the azimuth and distance to the wall, and generates a grid image indicating the contour of the surface of the surrounding wall facing the laser distance sensor. Then, the grid image is compared with the grid image of the map information indicating the contour of the wall in the moving area stored in advance, and overlaps with the grid image of the detected wall contour in the map information grid image. Find the part. Then, the current position on the map information is calculated based on the detected azimuth and distance to the surrounding wall, and the self position in the moving area is estimated. The distance sensor included in the autonomous traveling device described in Patent Document 1 scans the irradiation direction of the laser light in the horizontal direction, measures the distance to the object on a two-dimensional plane, and faces the distance sensor of surrounding objects. The contour of the part can be detected and is fixed to the autonomous mobile device.

However, the outline of an object in the movement area of the autonomous mobile device is generally different depending on the position in the height direction. Depending on the object, it is easy to collate the contour detected by the distance sensor with the contour stored in the map information at the contour at a certain height, but it is difficult to collate the contour at another height. There is a case.
And if the height at which the distance sensor is arranged in the autonomous mobile device is such a height that makes it difficult to collate the surrounding objects, the self-position estimation accuracy is lowered and the self-position may be lost.

  In order to solve the above-described problem, the present invention includes a moving unit, a map storage unit that stores map information of a moving region, a distance measuring unit that measures a distance between an object existing in the moving region, Estimating means for estimating self-position in the moving area by collating contour information of surrounding objects obtained from the measurement result of the distance measuring means with the contour information of objects in the moving area included in the map information; In the autonomous mobile device provided, the height of the distance measuring means can be changed, and the map storage means includes contour information of an object in the moving area corresponding to the height of the distance measuring means as the map information. It is characterized by that.

  According to the present invention, it is possible to improve the estimation accuracy of the self position of the autonomous mobile device.

Explanatory drawing which shows the map information which the map memory | storage part of a self-propelled robot memorize | stores, and the position of the height direction of the ranging sensor in each area. Explanatory drawing of the self-propelled robot of embodiment, (a) is a top view, (b) is a right view. The block diagram which shows an example of the control system of a self-propelled robot. The schematic diagram explaining the estimation method of the self position of the self-propelled robot. An explanatory view of an example of two kinds of shelves constituting a layout of a movement area of self-running robot 1 and a self-running robot system in which a self-running robot is arranged in the movement area. The flowchart of the position estimation loop control repeated from the start of autonomous running to the end of autonomous running. Explanatory drawing of the sensor base of a sensor raising / lowering mechanism, (a) is explanatory drawing of an automatic sensor base, (b) is explanatory drawing of a manual sensor base. Explanatory drawing of the other example of a sensor raising / lowering mechanism. Explanatory drawing of the self-propelled robot of the modification 1, (a) is a top view, FIG.5 (b) is a right view. Explanatory drawing of the self-propelled robot of the modification 2, (a) is a top view, (b) is a right view. Explanatory drawing of the self-propelled robot of the modification 3, (a) is a side view of a self-propelled robot, FIG. (B) is explanatory drawing of a pallet.

Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings.
2A and 2B are explanatory diagrams of the self-propelled robot 1 that is the autonomous traveling device of the embodiment. FIG. 2A is a top view of the self-propelled robot 1, and FIG. FIG. As an autonomous traveling device, there are a type in which an object is loaded on a vehicle and an object is unmannedly delivered to a designated place, and a type in which a cart is pulled. The self-running robot 1 of the present embodiment can be applied to any type of autonomous running device. The right side in FIG. 2 is the front of the self-propelled robot 1.

FIG. 3 is a block diagram illustrating an example of a control system for the self-running robot 1.
As shown in FIGS. 2 and 3, the self-propelled robot 1 includes a vehicle body 2, a drive wheel 3, an auxiliary wheel 4, a drive motor 13, a range sensor 5, and an encoder 16 for detecting the rotational speed of the drive wheel 3. And the arithmetic processing part 10 which performs autonomous running control is provided. The self-running robot 1 also includes a movement control unit 12 that controls the drive motor 13 and a map storage unit 11 that stores map information of the movement region. The movement control unit 12 creates a control signal for driving the drive wheels 3 and transmits the control signal to the drive motor 13.

  The sensor position control unit 14 in FIG. 3 controls the height of the range sensor 5 in the self-running robot 1 by controlling the driving of a sensor height adjusting motor described later. In addition, the work control unit 15 controls the operation of automatic work performed by the self-running robot 1 other than autonomous running, such as loading and unloading work when stopped.

  As shown in FIG. 2, the self-propelled robot 1 is provided with a non-contact range sensor 5 for recognizing an obstacle or the like that appears in the direction of moving to the front surface of the vehicle body 2. The range sensor 5 measures the distance to an object existing in the moving area and generates environment data. The arithmetic processing unit 10 compares the contour data of the surrounding objects obtained from the measurement result of the range sensor 5 with the contour data included in the map information stored in the map storage unit 11, so that the self-running robot 1 It has a function as estimation means for estimating the position.

In the present embodiment, a general two-dimensional laser range sensor (laser range finder) that irradiates the laser beam L is used as the range sensor 5 and is arranged so that the irradiation direction of the laser beam L is substantially horizontal. Has been.
The range sensor 5 irradiates the laser beam L while continuously changing the irradiation direction, and receives the reflected light from the object in the fan-shaped detection region, thereby measuring the distance to the object. be able to. In addition, the self-propelled robot 1 is disposed so that the front direction which is the center of the range sensor 5 and the scanning range coincides with the traveling direction when the self-propelled robot 1 travels straight.
As the range sensor 5, it is also possible to use an array that uses millimeter waves or ultrasonic waves, or a sensor that obtains data of the moving region by rotating the ultrasonic sensor.
In autonomous traveling control, the map information stored in the map storage unit 11 in advance, the travel distance estimated by odometry (the travel distance is calculated from the rotation speed of the encoder 16), and the distance detected by the range sensor 5 The self-position is estimated by matching the information.

  The map information stored in the map storage unit 11 is created based on the distance information measured by the range sensor 5 by moving the self-running robot 1 within the moving area before the operation of the self-running robot 1 starts. You may memorize | store in the map memory | storage part 11. FIG. Specifically, while the self-propelled robot 1 is moved, the range sensor 5 detects the azimuth and distance to the surrounding objects, and the detected object at the height at which the range sensor 5 is arranged. A grid image showing the contours of the opposing surfaces is generated. Then, the feature points of the grid image are collated before and after the movement, and the images are overlapped so that the same feature points are best matched, and map information of the entire movement region is created and stored in the map storage unit 11. As a method of moving the self-running robot 1 within the movement area when creating map information, it may be performed manually such as a human pressing, or the self-running robot 1 may be moved automatically.

  When the self-propelled robot 1 travels autonomously, the range sensor 5 detects the azimuth and distance to the object, and faces the range sensor 5 at the height at which the range sensor 5 is arranged for the detected surrounding objects. A grid image showing the contour of the surface to be generated is generated. Then, this grid image is collated with a grid image of map information indicating the outline of the fixed obstacle in the entire moving area stored in advance in the map storage unit 11, and Then, a portion that overlaps the grid image of the detected contour of the object is searched for. Then, the current position on the map information is calculated based on the detected azimuth and distance to the surrounding objects, and the self position in the moving area is estimated.

  FIG. 4 is a schematic diagram for explaining a method for estimating the self position of the self-running robot 1 in the moving region. 4A shows the positional relationship between the self-propelled robot 1 and the fixed obstacle 30 constituting the layout of the moving area, and FIG. 4B shows the position of the fixed obstacle 30 detected by the range sensor 5. FIG. 4C shows that the self-position is specified by matching.

  The position and shape of the fixed obstacle 30 in the movement area shown in FIG. 4A are stored in advance in the map storage unit 11 as map information. As shown in FIG. 4B, the contour of the fixed obstacle 30 is detected by the range sensor 5, and the self-position is estimated as shown in FIG. 4C by collating the detection result with the map information.

  FIG. 5 is an explanatory diagram of an example of two types of shelves constituting the layout of the movement area of the self-running robot 1 and the self-running robot system 100 in which the self-running robot 1 is arranged in the movement area. 5A is a front view of the two-tier shelf 40 and the three-tier shelf 50, and FIG. 5B is a two-dimensional view of the self-propelled robot system 100 at the height “H1” in FIG. It is explanatory drawing of the map information of a plane. As shown in FIG. 5B, the movement area of the self-propelled robot system 100 is divided into a first area 400 where only the two-tier shelf 40 is arranged and a second area 500 where only the three-tier shelf 50 is arranged. You can divide the area.

The movement area of the self-propelled robot 1 in the self-propelled robot system 100 of this embodiment shown in FIG. 5 is a warehouse for storing the products 90, and shelves (40, 50) for organizing the products 90 are arranged. Yes. As shown in FIG. 5, the two-stage shelf 40 includes a two-stage column part 42 and a two-stage shelf board 41, and the three-stage shelf 50 includes a three-stage column part 52 and a three-stage shelf board 51.
In a moving area such as a warehouse where the self-propelled robot 1 is operated, there are few environments composed only of obstacles whose contours are constant at any height.

  The contours of the second shelf 40 and the third shelf 50 are greatly different between the height at which the shelf plates (41, 51) are arranged and the height of only the pillar portions (42, 52). When the object is detected by the range sensor, if the area occupied by the object in the two-dimensional plane to be detected is large, such as the shelf boards (41, 51), the number of feature points for detecting the contour of the object increases. It is easy to perform matching between shape information and map information. On the other hand, when the area occupied by the object in the two-dimensional plane to be detected is small, such as only the pillars (42, 52), the number of feature points for detecting the contour of the object is reduced, and the detected contour and the contour of the object in the map information are reduced. It becomes difficult to perform matching with.

When the height of the range sensor 5 is “H1”, the contour of the second shelf 41 can be detected in the first area 400 because the second shelf 40 is the height at which the second shelf 41 is located. Since there are many feature points, matching with map information can be performed satisfactorily.
On the other hand, in the three-tier shelf 50, it is not the height at which the three-tier shelf 51 is located, but the height at which only the three-tier column portion 52 is located. For this reason, in the second zone 500, only the contour of the three-stage column portion 52 of the three-level shelf 50 can be detected, the number of feature points is small, and the matching information with the map information is insufficient, and the self-position estimation accuracy is insufficient. May lose its position and become inoperable.

The self-propelled robot 1 of the present embodiment includes a sensor lifting mechanism 8 having a sensor bracket 6 that holds a range sensor 5 and a sensor base 7 that can change the installation height of the sensor bracket 6. As indicated by the arrow “α”, the height of the range sensor 5 can be changed.
Further, the map storage unit 11 associates with each of a plurality of areas, and has a height at which the detection result of the range sensor 5 and the map information can be easily matched, that is, a height that is a two-dimensional plane with many feature points. The map information is stored.
When the self-propelled robot 1 travels in each area, the height of the range sensor 5 is set to a height corresponding to the map information of the two-dimensional plane having many feature points, and the measurement result is used. Estimate self position.

FIG. 1 is an explanatory diagram showing map information stored in the map storage unit 11 of the self-running robot 1 of the present embodiment and the position of the range sensor 5 in the height direction in each zone. FIG. 1A is an explanatory diagram of an example of map information stored in the map storage unit 11. For the first area 400, the map information of the two-dimensional plane at the height “H1” where the two-stage shelf 41 is arranged is stored, and for the second area 500, the three-stage shelf 51 is arranged. The map information of the two-dimensional plane at the height “H2” is stored.
FIG. 1B shows the position in the height direction of the range sensor 5 when the self-running robot 1 travels in the first area 400, and FIG. The position in the height direction of the range sensor 5 when traveling in the zone 500 is shown.

When the self-propelled robot 1 travels in the first zone 400, the height of the range sensor 5 is set to “H1”, and the height “H1” of the first zone 400 shown in the left diagram of FIG. The self-position is estimated based on the map information of the two-dimensional plane in “and the detection result of the range sensor 5”. When traveling in the second zone 500, the height of the range sensor 5 is set to “H2”, and the two-dimensional height at the height “H2” of the second zone 500 shown in the right diagram in FIG. The self position is estimated based on the map information of the plane and the detection result of the range sensor 5.
As described above, map information having a height with many feature points is prepared according to the area to be traveled, and the range sensor 5 is autonomously traveled according to the height corresponding to the map information.

In the self-propelled robot 1, the area on the map information is divided for each fixed obstacle having the same height at which the feature points can be easily found in the moving area to be traveled. And in each area, the map information of predetermined height with many feature points is used, and the height of the range sensor 5 is also adjusted to this predetermined height. Then, it autonomously travels by estimating its own position while collating the detection result of the range sensor 5 with the map information.
Thereby, the self-running robot 1 can improve the estimation accuracy of the self-position when the two-dimensional plane shape travels within a moving area where fixed obstacles differ depending on the height.

As a method for detecting a distance to an obstacle with a different height using a distance sensor arranged at a certain height, a method of changing the inclination of the irradiation direction of the laser light of the distance sensor is conceivable. However, in this method, a blind spot is generated on the near side of the laser beam irradiated obliquely.
On the other hand, in the self-running robot 1, the height of the range sensor 5 is adjusted to a predetermined height at which the feature points of each area can be easily found by changing the height of the range sensor 5 which is a distance sensor. It is possible to prevent blind spots from being formed at a predetermined height at which feature points can be easily found.

FIG. 6 is a flowchart of position estimation loop control that is repeated from the start of autonomous running of the self-running robot 1 to the end of autonomous running.
During autonomous traveling, obstacle data is acquired by the range sensor 5 (S1), and the self-position is estimated by matching with map information (S2). Based on the estimated self position, it is determined whether or not the area to which the self position belongs has changed from one of the first area 400 and the second area 500 to the other (S3). When the area has changed (“Yes” in “S3”), the map information of the predetermined height of the corresponding area is called (S4), and the height of the range sensor 5 is changed to the predetermined height (S5). Repeat the position estimation loop. When the area has not changed (“No” in “S3”), the position estimation loop is repeated without changing the height of the range sensor 5.

  As shown in FIG. 6, when a predetermined height with many feature points enters a different area by estimating the self-location, the map information at the predetermined height of the corresponding area is called and the height of the range sensor 5 is displayed. Is automatically adjusted to estimate the self-position.

Next, a method for creating map information on a two-dimensional plane having a large number of feature points in each area will be described. As described above, when the self-propelled robot 1 is moved within the movement area before the operation of the self-propelled robot 1 is started, the height of the range sensor 5 is set to a height with many feature points in each area. Move. This makes it possible to measure the distance to fixed obstacles existing in each area at a height where there are many feature points in each area, and to create two-dimensional plane map information with a high number of feature points in each area can do. Further, since the map information is created based on the feature points actually measured by the range sensor 5, it becomes easy to collate the range sensor 5 with the measurement result and the map information during autonomous traveling.
Not only the method of actually measuring by moving the self-propelled robot 1 within the moving area, but if there is electronic data related to the arrangement of the fixed obstacle in the moving area, the feature points of each area are based on this electronic data. The map information of a two-dimensional plane with a large height may be created.

  In the above-described embodiment, the configuration in which two types of shelves (40, 50) having different heights of the shelf boards (41, 51) are arranged in the two sections (400, 500), respectively, has been described. A plurality of types of obstacles having different heights as characteristic contours between a plurality of areas are not limited to shelves. For example, the contour of the desk differs between the height at which the top plate is positioned and the height at which only the feet that support the top plate are positioned. For this reason, the self-propelled robot 1 of the present embodiment can be arranged even in a moving area where desks with different top plates and chairs with different seating heights are arranged depending on the area.

  The configuration has been described in which the map information having the height “H1” is stored for the first section 400 and the map information having the height “H2” is stored for the second section 500. As map information memorize | stored in the map memory | storage part 11, it is good also as a structure which memorize | stores the map information in several height, respectively about several areas.

Moreover, it is good also as a structure which memorize | stores the map information in several height as map information of not only the thing which divides | segments a movement area | region into a some area but has map information but the whole movement area | region.
For example, in an office environment, a desk and a chair are arranged. The chair has a different contour between the height at which the seating surface is located and the height at which only the feet supporting the seating surface are located. In general, the top of the desk is higher than the seat surface of the chair. Therefore, as the map information, two-dimensional plane map information at the height of the table top and two-dimensional plane map information at the height of the seat surface of the chair are stored. And at the time of autonomous running start, when autonomous driving is performed with the height of the range sensor 5 as the height of the top of the desk, and based on the map information and odometry, it is determined that the chair is located around May be configured to change the height of the range sensor 5 to the height of the seat surface and perform autonomous traveling.

  In this way, a configuration is adopted in which the height of the range sensor 5 is changed when it is determined that the height of many feature points in surrounding obstacles has changed based on the map information and odometry without dividing the area. The possible movement area is not limited to the office environment. For example, shelves with different heights of shelves are not divided into areas and can be applied in a mixed warehouse environment.

  FIG. 7 is an explanatory diagram of the sensor base 7 of the sensor lifting mechanism 8. FIG. 7A is an explanatory diagram of the sensor base 7 that automatically raises and lowers the sensor bracket 6, and FIG. 7B is an explanatory diagram of the sensor base 7 that manually changes the vertical position of the sensor bracket 6. It is.

  The sensor base 7 in FIG. 7A is a slide table including a guide 7a and a moving table 7b. The sensor base 7 includes a sensor height adjustment motor, and the vertical position of the moving table 7b relative to the guide 7a is changed by driving the sensor height adjustment motor. And the sensor bracket 6 is being fixed to this movement table 7b, and the height of the sensor bracket 6 changes because the sensor position control part 14 controls the drive of a sensor height adjustment motor. Thereby, when the area where the self-running robot 1 travels changes, the height of the range sensor 5 held by the sensor bracket 6 can be automatically adjusted. By automatically adjusting the height of the range sensor 5, it is possible to prevent time for adjusting the height of the range sensor 5 from being taken when going between a plurality of areas.

The sensor base 7 in FIG. 7B is configured to include a positioning boss hole 7c. By providing a positioning pin in the sensor bracket 6 and changing the height of the boss hole 7c into which the positioning pin is inserted, the height of the range sensor 5 held by the sensor bracket 6 can be adjusted.
The configuration in which the height is manually adjusted is that when the area where the self-propelled robot 1 travels changes, the self-propelled robot 1 stops and the range sensor 5 is set to a predetermined height in the area where the next traveling is scheduled. Adjust the height manually and resume autonomous driving.

The self-propelled robot system 100 is not limited to a configuration in which a single self-propelled robot 1 travels between the first area 400 and the second area 500 having many feature points and different heights. The self-propelled robot 1 may be arranged in each of the first area 400 and the second area 500.
In this case, the self-propelled robot 1 arranged in the first area 400 sets the height of the range sensor 5 to “H1”, and the map information stored in the map storage unit 11 includes the height “H1” of the first area 400. It is good also as only the map information of the two-dimensional plane in. Similarly, the self-propelled robot 1 arranged in the second section 500 sets the height of the range sensor 5 to “H2”, and the map information stored in the map storage unit 11 is the height “H2” of the second section 500. Only map information on a two-dimensional plane.

Thus, in the structure which arrange | positions the self-propelled robot 1 for every area, since the height of the range sensor 5 can be changed, of the self-propelled robot 1 arrange | positioned in each area where height with many feature points differs. Sharing can be achieved, and procurement costs can be reduced.
Moreover, when arrange | positioning the self-propelled robot 1 for every area, the frequency which changes the height of the range sensor 5 is low. For this reason, the height of the range sensor 5 is automatically adjusted by manually changing the height of the range sensor 5 like the sensor lifting mechanism 8 having the sensor base 7 shown in FIG. The cost per unit of the self-propelled robot 1 can be reduced as compared with the configuration that can be changed by the above.

FIG. 8 is an explanatory diagram of another example of the sensor lifting mechanism 8.
The sensor base 7 of the sensor elevating mechanism 8 shown in FIG. 8 includes a moving stage portion 7d, a linear stage 7e, a screw shaft 7g, and a screw shaft rotating motor 7f. By controlling the rotational drive of the screw shaft rotation motor 7f, the moving stage portion 7d to which the sensor bracket 6 is fixed moves in the vertical direction, and the height of the range sensor 5 held by the sensor bracket 6 is automatically adjusted. It becomes possible.

[Modification 1]
Next, a first modified example (hereinafter referred to as “modified example 1”) of the self-running robot 1 will be described.
FIG. 9 is an explanatory diagram of the self-running robot 1 according to the first modification. FIG. 9A is a top view of the self-running robot 1 and FIG. 9B is a right side view of the self-running robot 1. .

The self-propelled robot 1 of Modification 1 includes three range sensors, one central range sensor 5a and two end range sensors 5b, as the range sensors 5. One central range sensor 5a is attached to a sensor base 7 that can automatically move up and down.
The use of a plurality of range sensors 5 increases the number of feature points to be measured, facilitates matching between the detected shape information and map information, and improves self-position estimation accuracy.

[Modification 2]
Next, a second modification of the self-running robot 1 (hereinafter referred to as “Modification 2”) will be described.
FIG. 10 is an explanatory diagram of the self-propelled robot 1 according to the second modification. FIG. 10 (a) is a top view of the self-propelled robot 1, and FIG. 10 (b) is a right side view.

The self-propelled robot 1 of Modification 2 has a configuration in which the range sensor 5 is installed on a multi-axis robot arm 27.
As indicated by an arrow “α” in FIG. 10B, the range of the range sensor 5 can be measured without providing a dedicated lifting means by controlling the tip of the robot arm 27 to move in the vertical direction. It is possible to change the vertical position of the sensor 5.
Further, by providing the range sensor 5 on the robot arm 27, the position of the range sensor 5 can be adjusted not only in the vertical direction but also in the horizontal direction, the front-rear direction, and the rotation direction.
When autonomous driving is performed, in one area, the vertical position of the range sensor 5 is kept constant so that an appropriate self can be obtained based on map information at a predetermined height with many feature points in the area. Position estimation can be performed.

  Further, by providing the robot arm 27 with a hand or a tool, the arm operation is performed when the self-propelled robot 1 is stopped. The range sensor 5 is used as a positioning sensor for measuring the distance between a surrounding object and the robot arm 27 when performing the arm operation. As described above, since the information measured by the range sensor 5 is used for operation control of the robot arm 27, the range sensor 5 can be used as a sensor for arm work. There is no need to provide a separate sensor. As a result, the number of sensors can be reduced, and the cost can be reduced.

Further, even if a distance sensor that cannot detect the surrounding 360 [°] is used as the range sensor 5, it is possible to obtain environmental data of the surrounding 360 [°] by changing the direction of the robot arm 27. Self-position estimation accuracy is improved.
Further, by directing the tip of the robot arm 27 toward the rear side of the self-propelled robot 1, an obstacle can be detected in the same way as when moving forward, and the robot can operate in the same way as when moving forward.

In addition, when the vehicle reaches a place with poor visibility such as a corner in the moving area, the robot arm 27 controls the position of the range sensor 5 to move to the front of the self-propelled robot 1 on the exit side of the corner. May be. By moving the range sensor 5 to the exit side of the corner, it is possible to accurately detect the obstacle in the left and right direction of the corner that becomes a blind spot on the front side of the corner.
At this time, even if the height of the range sensor 5 is different from the predetermined height of the traveling area, the exit at the corner may be preferentially detected. In this case, after detecting the corner exit, the robot arm 27 is controlled so that the height of the range sensor 5 is returned to the predetermined height of the area.
Furthermore, by projecting the position of the range sensor 5 with respect to the main body of the self-propelled robot 1 by the robot arm 27 in the front, rear, left, and right directions, it becomes possible to reduce the blind spot due to the main body of the device itself.

[Modification 3]
Next, a third modification of the self-running robot 1 (hereinafter referred to as “modification 3”) will be described.
FIG. 11 is an explanatory diagram of the self-propelled robot 1 according to the modification 3. FIG. 11 (a) is a side view of the self-propelled robot 1, and FIG. 11 (b) is a diagram of the pallet 70 in FIG. 11 (a). It is explanatory drawing.

The self-propelled robot 1 of Modification 3 is a forklift, and as shown by an arrow “α” in FIG. 11A, the range sensor 5 is installed in a lift device 37 that moves the fork 37a up and down.
The lift device 37 is a device that raises and lowers the fork 37a with respect to the mast 37b, and uses a function of lifting an article such as a pallet 70 using the fork 37a.

In the self-propelled robot 1 of the third modification, by controlling the fork 37a of the lift device 37 to move in the vertical direction, the vertical direction of the range sensor 5 is not provided without providing a dedicated lifting means for the range sensor 5. The position of can be changed.
Further, when the fork 37a of the self-propelled robot 1 which is a forklift is inserted into the insertion port 70a of the pallet 70, the insertion port 70a of the pallet 70 is sensed using the information of the range sensor 5 and inserted. Work can be done.
In this way, the range sensor 5 can be used as an autonomous traveling sensor during movement and as a positioning sensor during lift work, and it is not necessary to provide each sensor individually, thereby reducing costs.

  What was demonstrated above is an example, and there exists an effect peculiar for every following aspect.

(Aspect A)
Moving means such as the drive motor 13 and the drive wheel 3, map storage means such as the map storage unit 11 that stores map information of the moving area such as a warehouse, the two-tier shelf 40 and the three-tier shelf 50 existing in the moving area A distance measuring means such as a range sensor 5 that measures the distance to an object such as a distance sensor, and the contour information of surrounding objects obtained from the measurement result of the distance measuring means are collated with the contour information of the object in the moving area included in the map information Thus, in the autonomous mobile device such as the self-propelled robot 1 provided with the estimation means such as the arithmetic processing unit 10 for estimating the self-position in the movement area, the height of the distance measurement means can be changed, and the map storage means Is provided with contour information (such as map information of a two-dimensional plane having a predetermined height) in the moving area corresponding to the height of the distance measuring means as map information.
According to this, as described in the above embodiment, the distance measuring unit is set to a height at which the contour information of the surrounding object detected by the distance measuring unit can be easily compared with the contour information stored in the map information. It is possible to change the height. Then, the autonomous mobile device estimates the self-position based on the contour information of the object in the moving area in the map storage means corresponding to the height and the contour information of the object obtained from the measurement result of the distance measuring means. It is possible to improve the estimation accuracy of the self-position.

(Aspect B)
In the aspect A, the map storage means such as the map storage unit 11 has predetermined heights such as heights “H1” and “H2” in which the contours of the objects such as the second shelf 40 and the third shelf 50 in the movement region are characteristic. The moving area is divided into a plurality of areas such as the first area 400 and the second area 500 according to the height of the map, and map information on a two-dimensional plane at a predetermined height is used as the map information for each of the plurality of areas. And the height of the distance measuring means such as the range sensor 5 is adjusted to a predetermined height in the area to be traveled, and the self-position in the moving area is estimated by the estimating means such as the arithmetic processing unit 10. Do.
According to this, as described in the above embodiment, self-position estimation is performed by using a map of a two-dimensional plane at a height where there are many feature points in the contour of an object in each area. The position estimation accuracy is improved and stable movement is possible.

(Aspect C)
In the aspect B, the map information of the two-dimensional plane at a predetermined height such as the heights “H1” and “H2” is measured for each of a plurality of areas such as the first area 400 and the second area 500. This is map information obtained by arranging distance measuring means such as the sensor 5 at a predetermined height and measuring the distance from the object such as the second shelf 40 and the third shelf 50 existing in the area.
According to this, as described in the above embodiment, it becomes easy to collate the range finding means, the measurement result, and the map information during autonomous traveling, and the self-position estimation accuracy is improved.

(Aspect D)
In any one of the aspects A to C, a plurality of ranging means such as the ranging sensor 5 are provided, and the height of at least one of the plurality of ranging means (such as the central ranging sensor 5a) can be changed.
According to this, as described in the first modification, the number of feature points to be measured increases by using a plurality of distance measuring means, and the detected object contour information is collated with the object contour information in the map information. This makes it easier to perform the self-position estimation accuracy.

(Aspect E)
In any one of the aspects A to D, an elevating unit such as a slide table that automatically elevates the distance measuring unit such as the range sensor 5 is provided.
According to this, as described in the above-described embodiment, the time for adjusting the height of the distance measuring means when automatically moving between a plurality of areas by automatically adjusting the height of the distance measuring means. Can be prevented.

(Aspect F)
In aspect E, the elevating means is constituted by a multi-axis robot arm such as the robot arm 27.
According to this, as described in the second modification, it is possible to change the vertical position of the distance measuring means without providing a dedicated lifting means for the distance measuring means such as the range sensor 5. . Moreover, since the blind spot can be reduced by moving the distance measuring means not only in the height direction but also in the left and right and rotation directions, self-position estimation accuracy and obstacle avoidance performance are improved.

(Aspect G)
In the aspect F, information measured by distance measuring means such as the range sensor 5 is used for operation control of the robot arm such as the robot arm 27.
According to this, as described in the second modification, one distance measuring unit can be used for both obstacle detection during movement and positioning during arm work, thereby reducing costs. .

(Aspect H)
In aspect E, the elevating means is configured by a lift device such as a lift device 37 that elevates and lowers the load such as the pallet 70.
According to this, as described in the second modification, it is possible to change the vertical position of the distance measuring means without providing a dedicated lifting means for the distance measuring means such as the range sensor 5. . In addition, one distance measuring means can be used both for detecting obstacles during movement and for positioning during lift work, thereby reducing the cost.

(Aspect I)
An autonomous mobile device such as a self-propelled robot 1 and a moving area such as a warehouse to which the autonomous moving device can move are provided. The autonomous moving device includes moving means such as a drive motor 13 and driving wheels 3 and map information of the moving area. A map storage means such as the map storage unit 11 for storing the distance, a distance measuring means such as a range sensor 5 for measuring the distance between an object such as the second shelf 40 and the third shelf 50 existing in the moving area, Estimating means such as the arithmetic processing unit 10 for estimating the self-position in the moving region by collating the contour information of the surrounding object obtained from the measurement result of the distance unit with the contour information of the object in the moving region included in the map information. In the autonomous mobile device system such as the self-propelled robot system 100 having the configuration, the moving region has a predetermined height such as heights “H1” and “H2” at which the contours of the objects arranged therein are characteristic. First zone 4 with different The map storage means has map information of a two-dimensional plane at a predetermined height for each of the plurality of areas, and the autonomous mobile device has a plurality of areas such as 0 and the second area 500 The height of the distance measuring means configured to be changeable is set to a predetermined height in the area to be traveled, and the self-position within the moving area is estimated by the estimating means.
According to this, as described in the above embodiment, the self-position estimation accuracy is improved by performing self-position estimation using a two-dimensional plane map at a height with many feature points in each area. , Stable movement becomes possible.

DESCRIPTION OF SYMBOLS 1 Self-propelled robot 2 Vehicle main body 3 Driving wheel 4 Auxiliary wheel 5 Range sensor 5a Center range sensor 5b End range sensor 6 Sensor bracket 7 Sensor base 7a Guide 7b Moving table 7c Boss hole 7d Moving stage unit 7e Linear stage 7f Screw shaft rotation motor 7g Screw shaft 8 Sensor lifting mechanism 10 Arithmetic processing unit 11 Map storage unit 12 Movement control unit 13 Drive motor 14 Sensor position control unit 15 Work control unit 16 Encoder 27 Robot arm 30 Fixed obstacle 37 Lift device 37a Fork 37b Mast 40 Second-tier shelf 41 Second-tier shelf plate 42 Second-tier column 50 Third-tier shelf 51 Third-tier shelf plate 52 Third-tier column 70 Pallet 70a Insert 90 Product 100 Self-propelled robot system 400 First zone 500 Second zone L Laser light

JP 2011-65202 A

Claims (9)

Transportation means;
Map storage means for storing map information of the moving area;
Distance measuring means for measuring a distance to an object existing in the moving area;
Estimating means for estimating a self-position in the moving area by comparing contour information of surrounding objects obtained from the measurement result of the distance measuring means with outline information of objects in the moving area included in the map information; In an autonomous mobile device comprising:
The height of the distance measuring means can be changed,
The autonomous mobile device, wherein the map storage means includes contour information of an object in the moving area corresponding to the height of the distance measuring means as the map information. The autonomous mobile device according to claim 1,
The map storage means divides the moving area into a plurality of sections according to a predetermined height at which an outline of an object in the moving area becomes characteristic, and the predetermined height for each of the plurality of sections. Storing the map information of the two-dimensional plane in as the map information,
An autonomous mobile device characterized in that the height of the distance measuring means is adjusted to the predetermined height in an area to be traveled, and the self-position is estimated in the moving area by the estimating means. The autonomous mobile device according to claim 2,
The map information of the two-dimensional plane at the predetermined height is measured for each of the plurality of areas by arranging the distance measuring means at the predetermined height and measuring the distance to an object existing in the area. An autonomous mobile device characterized in that it is map information obtained by doing. The autonomous mobile device according to any one of claims 1 to 3,
An autonomous mobile device comprising a plurality of the distance measuring means, wherein at least one height of the plurality of distance measuring means can be changed. The autonomous mobile device according to any one of claims 1 to 4,
An autonomous mobile device comprising lifting means for automatically raising and lowering the distance measuring means. The autonomous mobile device according to claim 5,
The autonomous moving apparatus according to claim 1, wherein the elevating means comprises a multi-axis robot arm. The autonomous mobile device according to claim 6,
An autonomous mobile device characterized in that information measured by the distance measuring means is used for operation control of the robot arm. The autonomous mobile device according to claim 5,
The autonomous moving device according to claim 1, wherein the lifting means is constituted by a lift device that lifts and lowers a load. An autonomous mobile device;
A moving area in which the autonomous mobile device is movable,
The autonomous mobile device includes a moving means, a map storage means for storing map information of the moving area, a distance measuring means for measuring a distance from an object existing in the moving area, and a measurement result of the distance measuring means. An autonomous mobile device having an estimation means for estimating self-position in the moving region by comparing contour information of surrounding objects obtained from the above with contour information of objects in the moving region included in the map information In the system,
The moving area has a plurality of areas having different predetermined heights where the outline of an object disposed therein is characteristic;
The map storage means has, as the map information, two-dimensional plane map information at the predetermined height for each of the plurality of areas,
The autonomous mobile device adjusts the height of the distance measuring means configured to be changeable to the predetermined height in the area to be traveled, and estimates the self-position in the moving area by the estimating means. An autonomous mobile device system characterized by performing.

Images