JP2007156576A - Method and device for adjusting odometry(wheel range finder) parameter for traveling carrier - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the odometry (wheel range finder) parameter adjustment method and device of an autonomous traveling carrier for easily and accurately adjusting such a correction parameter as the manufacture error or assembly error of a wheel relating to traveling quantity and turning quantity per the revolution of a wheel and any error to be generated due to abrasion due to the lapse of a time in an autonomous traveling carrier such as a robot or a vehicle (AGV) autonomously traveling in a wheel form. <P>SOLUTION: A marker which can be imaged by an imaging device is installed in the neighborhood of a goal position in a specific distance from the start position of an autonomous traveling carrier, and the image of the marker which has been preliminarily imaged by the imaging device at the goal position is prepared as a template, and the autonomous traveling carrier is installed at the start position so as to be faced to the goal direction, and a specific distance traveling instruction is made, or one revolution turning command is given at the goal position, and the pickup image of the marker imaged after traveling or revolution and the preliminarily picked-up template image are used to adjust the odometry (wheel range finder) parameter. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an odometry (wheel distance meter) parameter adjusting method and apparatus for a traveling vehicle that adjusts parameters related to wheels in the traveling vehicle so that the traveling vehicle traveling autonomously accurately reaches a target position. Autonomous vehicle odometry (wheel distance meter) parameter adjustment method for accurately adjusting parameters such as travel amount per wheel rotation and turn amount in autonomous vehicle such as Beagle (AGV) that autonomously travels in the form of vehicles and vehicles And the apparatus.

  A beagle (AGV) that autonomously travels in the form of a robot or vehicle that is required to travel through a narrow space with various obstacles, such as indoors or factories, through a door, etc. ), Etc., to reach the target position accurately, it is necessary to travel accurately corresponding to the instructed travel distance and direction and to know its own position accurately. is there. Therefore, for example, in the case of two-wheel drive, it has an odometry function (movement distance measurement function) for estimating the azimuth of the movement amount of the traveling carriage based on the rotation amount of the wheel and the difference between the rotation amounts of the two wheels.

  In this odometry function, the wheel diameter and the distance (tread) between the two wheels are important parameters (hereinafter referred to as odometry parameters) that affect the accuracy in estimating the amount of movement and direction.

  However, in general, errors occur due to wheel manufacturing errors, assembly errors, and wheel wear over time.As a result, these parameters differ from the design values. Will cause an error. In particular, in a traveling vehicle for autonomous movement such as a beagle (AGV) that autonomously travels in the form of a robot or a vehicle, if the travel distance and direction in the actual travel space differ from the estimated values, the estimated travel route Since it deviates, it causes a failure of autonomous movement by causing a collision with an object outside the assumed route.

  For this reason, the above-described odometry parameters are corrected in consideration of manufacturing errors and assembly errors. Conventionally, such corrections are performed for the traveling vehicle with respect to a movement command for a specified distance and a turn command for a specified angle. The actual movement amount and turning amount are measured, and the wheel diameter and tread value are obtained from the error from the measured command value (reverse operation when estimating the movement distance and direction) manually.

  However, manual work requires a lot of time for adjustment, and there is a problem that the cost increase due to this adjustment work is inevitable. In addition, when the wheel diameter changes due to wear or the like with the passage of time, there is a problem that the initial odometry parameters cannot be used alone.

  Regarding a method for removing an error from the distance estimated value by such an odometer, for example, in Patent Document 1, in a place where a GPS (Global Positioning System) signal cannot be used like an urban area with an obstacle. A system that corrects the signal obtained from the odometer using the tire slip signal from the anti-lock brake system (ABS) and traction control system (TCS) to enable accurate position estimation. It is shown.

  Patent Document 2 includes means for imaging and recognizing a specific facility such as a kilometer post existing at regular intervals along the road in order to improve the detection accuracy of the vehicle position, and the travel distance is calculated for each recognition of the specific facility. In addition, when the travel distance approximates an integral multiple of the distance between specific facilities, it is added to the detection value of the travel distance, and when it does not approximate the integral multiple, it is not added so that the position of the vehicle is accurately detected. Is disclosed.

JP 2001-336950 A JP-A-9-119851

  However, the techniques disclosed in Patent Documents 1 and 2 are techniques for correcting the detection distance of an odometer in a vehicle that travels a long distance, such as an automobile, and a large accuracy is not required. There are many problems in terms of accuracy when applied to traveling carts that are required to travel accurately in narrow spaces and short distances such as indoors and factories, such as traveling beagles (AGV).

  In particular, in the methods disclosed in these Patent Documents 1 and 2, a tire slip signal from an antilock brake system (ABS) or a traction control system (TCS) is used to correct a signal obtained from an odometer ( (Patent Document 1) or capturing and recognizing a specific facility such as a kilometer post existing at regular intervals along the road, and calculating the travel distance for each recognition of the specific facility (Patent Document 2) to correct the odometer However, with such a method, it is impossible to accurately specify the rotation amount of the wheel or the difference in the rotation amount between the wheels.

  Therefore, in the present invention, in an autonomous traveling vehicle such as a beagle (AGV) that autonomously travels in the form of a robot or a vehicle, a wheel manufacturing error or assembly error related to a traveling amount or a turning amount per wheel rotation, and a time It is an object to provide an odometry (wheel distance meter) parameter adjustment method and apparatus for an autonomous traveling vehicle that can easily and accurately adjust correction parameters such as errors caused by wear due to progress.

In order to solve the above problems, the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle in the present invention is as follows.
In an autonomous traveling vehicle that travels with at least two traveling wheels each having an independent drive source and an imaging device, the travel distance is determined by the amount of rotation of the traveling wheel and the difference in the amount of rotation of each traveling wheel. Odometry function (movement distance measurement function) for estimating the moving direction, and an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating the moving distance and the moving direction A method for adjusting the odometry (wheel distance meter) parameter of an autonomous traveling vehicle,
After setting a marker that can be imaged by the imaging device in the vicinity of a goal position at a specific distance from the start position of the autonomous traveling carriage, and preparing an image of the marker previously captured by the imaging device at the goal position as a template The autonomously traveling carriage is installed at the start position in the goal direction and gives a travel instruction for the specific distance, or a turn command for one rotation is given at the goal position, and the marker imaged after running or after turning The odometry (wheel distance meter) parameter is adjusted using the captured image and the template image.

Moreover, the apparatus which implements the odometry (wheel distance meter) parameter adjustment method of this autonomous traveling vehicle is:
A wheel diameter and a wheel interval for estimating a moving distance and a moving direction by traveling with at least two traveling wheels and an imaging device each having an independent driving source and a means for measuring the rotation amount ( An odometry (wheel distance meter) parameter adjusting device for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by
Storage means for storing, as a template, a marker installed in the vicinity of a goal position at a specific distance from the start position of the autonomous traveling vehicle so as to be imaged by the imaging device, and an image of the marker previously captured by the imaging device at the goal position The autonomous traveling carriage is installed at the start position in the goal direction to give a travel instruction for the specific distance, or a turn command for one rotation is given at the goal position, and after the travel or turn, the imaging device Imaging the marker and measuring the rotation measured by the output of the rotation amount measuring means, or the difference between the rotation amount of the traveling wheel in traveling and the rotation amount of each traveling wheel, the template stored in the storage device, and the turning Alternatively, the travel distance and travel direction of the autonomous traveling vehicle are estimated from the comparison result with the marker image captured after traveling. Characterized in that a control means for adjusting the odometry (wheel rangefinder) parameters.

Similarly,
The amount of rotation of the traveling wheel and the amount of rotation of each traveling wheel in an autonomous traveling vehicle traveling with at least two traveling wheels having independent drive sources and front and side obstacle detection sensors. An odometry function (movement distance measurement function) for estimating a movement distance and a movement direction based on the difference, and an odometry determined by a wheel diameter and a wheel interval (tread) for estimating the movement distance and the movement direction. A method for adjusting an odometry (wheel distance meter) parameter of an autonomous traveling vehicle, which adjusts a (wheel distance meter) parameter,
A measuring jig having a substantially U-shaped flat plate is prepared and installed so that the distance from the starting position of the autonomous traveling carriage to each flat plate constituting the measuring jig is a predetermined distance. After setting the traveling carriage toward the goal position separated by a specific distance, after calculating the distance from the front and side obstacle detection sensors to each flat plate in the measurement jig in advance,
When measuring the turning angle error, the turning command for one rotation is directly used. When calculating the running error, the measuring jig is installed at the goal position to give the running command for the specific distance, and the front after turning or after running And the distance to each flat plate of the measurement jig by the side obstacle detection sensor is calculated, and the odometry (wheel distance meter) parameter of the autonomous traveling vehicle is calculated from the difference from the value measured before the turning or the traveling. It is characterized by adjusting.

Moreover, the apparatus which implements the odometry (wheel distance meter) parameter adjustment method of this autonomous traveling vehicle is:
To travel with at least two traveling wheels each having an independent drive source and a means for measuring the amount of rotation and sensors for detecting front and side obstacles, and to estimate the travel distance and travel direction An odometry (wheel distance meter) parameter adjusting device for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) of the vehicle,
The distance from the autonomous traveling cart installed toward the goal position that is a specific distance away from the start position to each flat plate in the measurement jig is a predetermined distance. The measurement jig is installed so that the distance to each flat plate measured in advance by the front and side obstacle detection sensors is stored, and when the turning angle error is measured, one turn is turned as it is. When calculating a running error, the measurement jig is installed at the goal position to give a running command for the specific distance, and the measurement after turning or running by the front and side obstacle detection sensors. A control hand that measures the distance to each flat plate of the jig and adjusts the odometry (wheel distance meter) parameter of the autonomous traveling vehicle from the difference from the value measured before turning or traveling stored in the storage means Characterized by comprising and.

Similarly,
An autonomous traveling vehicle that has at least two traveling wheels each having an independent drive source and an obstacle detection sensor that includes at least two beams having an angular difference capable of measuring the inclination of the autonomous traveling vehicle. The odometry function (movement distance measurement function) for estimating the movement distance and the movement direction based on the rotation amount of the traveling wheel and the difference between the rotation amounts of the respective traveling wheels, the movement distance and the movement direction An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimation,
A flat plate measuring jig capable of reflecting the beam in the obstacle detection sensor is prepared, and the autonomous traveling carriage is installed at a start position at a goal position separated by a specific distance and the flat plate measuring jig is measured. After calculating the autonomous traveling carriage posture angle from the distance to the flat plate measurement jig and the irradiation angle and detection distance of each beam by each beam in the obstacle detection sensor in advance,
When measuring the turning angle error, the turn command for one rotation is used as it is, and when calculating the running error, the flat plate measuring jig is installed at the goal position to give the running command for the specific distance. The autonomous traveling vehicle attitude angle is calculated from the distance to the flat plate measuring jig by the respective beams in the obstacle detection sensor, the irradiation angle of each beam, and the detection distance, and from the difference from the value before the turn or before the travel. The odometry (wheel distance meter) parameter of the autonomous traveling vehicle is adjusted.

Moreover, the apparatus which implements the odometry (wheel distance meter) parameter adjustment method of this autonomous traveling vehicle is:
For obstacle detection comprising at least two beams having an angular difference capable of measuring the traveling direction inclination of at least two traveling wheels and autonomous traveling carts each having an independent drive source and a means for measuring the rotation amount Odometry (wheel) of an autonomous traveling vehicle that travels with a sensor and adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating a moving distance and a moving direction Distance meter) parameter adjusting device,
A flat plate measuring jig capable of reflecting the beam in the obstacle detection sensor, and a distance from the autonomous traveling carriage installed toward a goal position a specific distance away from the start position to the flat plate measuring jig is predetermined. The flat plate measuring jig is installed so that the distance is the same, and the detection distance calculated autonomously from the distance to the flat plate measuring jig and the irradiation angle of each beam by each beam in the obstacle detection sensor in advance A storage means for storing a traveling carriage attitude angle; a turn command for one rotation as it is at the time of turning angle error measurement; and a flat plate measuring jig installed at the goal position to calculate a running error; The detection distance calculated from the distance to the flat plate measuring jig and the irradiation angle of each beam after turning or traveling by the obstacle detection sensor and the autonomous traveling platform From the difference between the value measured before pre-turning or running stored in the attitude angle and the storage means, characterized in that a control means for adjusting the odometry (wheel rangefinder) parameters of the autonomous traveling carriage.

  Thus, when sensors such as an imaging device, a front and side obstacle detection sensor, and an obstacle detection sensor having at least two beams are provided in the autonomous traveling vehicle, these are newly used. Without adding a sensor, straight travel error and turning error can be calculated automatically, cost can be reduced, and odometry (wheel distance meter) parameters of the autonomous traveling vehicle can be calculated without human intervention. Can be adjusted. In addition, when the front and side obstacle detection sensors and the obstacle detection sensor having at least two beams are used, the linear movement error is more accurate than when using the image captured by the imaging device. The turning error can be calculated. Furthermore, in front and side obstacle detection sensors, it is necessary to manually position the autonomous traveling carriage with high accuracy with respect to a measuring jig formed in a substantially U-shaped flat plate, which is very laborious. In contrast, when an obstacle detection sensor having at least two beams is used, such a labor is not necessary because the measuring jig is a flat plate, and it is very straight forward. Movement error and turning error can be calculated.

Similarly,
In an autonomous traveling vehicle that travels with at least two traveling wheels each having an independent drive source, the travel distance and the travel direction are determined by the amount of rotation of the travel wheels and the difference in the amount of rotation of each travel wheel. An odometry function (movement distance measurement function) for estimating, and adjusting an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating the movement distance and the movement direction; An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle,
A position / orientation detection sensor for detecting the position and orientation of the autonomous traveling vehicle is provided in the traveling space of the autonomous traveling vehicle, and a substantially perpendicular posture measuring jig is attached to the autonomous traveling vehicle, so that the position / orientation detection sensor is installed. The autonomous traveling vehicle is installed at a position so that the distance from each surface of the jig for posture measurement to the sensor is a predetermined distance, and the position / posture is measured. When the command is given and the running error is measured, the autonomous traveling carriage is installed at a specific distance from the position / posture detection sensor so that each surface of the posture measuring jig faces a predetermined direction. A straight command is given, and the difference between the angle given by the turn command from the position / posture before and after turning or after running of the posture measuring jig detected by the position / posture detection sensor, or the straight command Given It obtains the difference between the specified distance, and adjusting the odometry (wheel rangefinder) parameters of the autonomous traveling carriage.

Moreover, the apparatus which implements the odometry (wheel distance meter) parameter adjustment method of this autonomous traveling vehicle is:
The vehicle has at least two traveling wheels each having an independent drive source and a means for measuring the rotation amount, and has a wheel diameter and a wheel interval (tread) for estimating a moving distance and a moving direction. An odometry (wheel distance meter) parameter adjustment device for an autonomous traveling vehicle that adjusts a determined odometry (wheel distance meter) parameter,
A position / posture detection sensor for detecting the position and posture of the autonomous traveling vehicle provided in the traveling space of the autonomous traveling vehicle, a substantially right-angle posture measuring jig attached to the autonomous traveling vehicle, and the position / posture detection Storage means for storing the result of measuring the position / orientation of the autonomous traveling carriage installed at a sensor position so that the distance from each surface of the attitude measuring jig to the sensor is a predetermined distance, and measuring a turning error When the vehicle travels, a turn command for one rotation is given as it is, and when the travel error is measured, the autonomous traveling vehicle is such that each surface of the posture measuring jig faces a predetermined direction at a specific distance from the position / posture detection sensor. Is installed to give a straight-ahead command for the specific distance, and the position of the autonomous traveling vehicle is determined based on the position / posture before and after turning of the posture measuring jig detected by the position / posture detection sensor. Distance is characterized in that a control means for adjusting the odometry (wheel rangefinder) parameters for estimating the moving direction.

  By doing so, it is possible to easily adjust the odometry (wheel distance meter) parameter even in an autonomous traveling vehicle that does not have an imaging device or a sensor.

  And in the adjustment method of the odometry parameter using the imaging device, the marker is composed of at least two markers, and the image of the marker previously captured by the imaging device at the goal position and the marker of the autonomous traveling vehicle The distance is used to determine the distance between the markers so that the distance between the autonomous traveling carriage and the marker after traveling can be calculated, or the marker is composed of at least two markers, and the distance between the markers is determined by the goal position. The distance between the marker image captured by the imaging device and the marker of the autonomous traveling vehicle is a preferred embodiment of the present invention.

  Further, the difference between the goal position and the actual arrival position of the autonomous traveling vehicle is calculated from the difference between the marker image captured after the traveling and the pre-captured template image, and the odometry parameter during straight traveling of the autonomous traveling vehicle is calculated. Or calculating the difference between the actual position of the autonomous traveling carriage and the goal position based on the difference in distance to each flat plate of the measuring jig measured after the starting position and after traveling, Calculate the odometry parameters when traveling straight ahead, and calculate the difference between the actual position of the autonomous traveling vehicle and the goal position based on the difference between the starting position and the distance to the flat plate measuring jig measured after traveling. Then, by calculating the odometry parameter for the straight traveling of the autonomous traveling vehicle, the odometry (wheel distance meter) parameter for the straight traveling can be easily obtained. It can be adjusted.

  Then, a turning angle error of the autonomous traveling vehicle is calculated from a difference between the marker image that is installed at the goal position and imaged after the turn and the pre-captured template image, and an odometry parameter when turning the autonomous traveling vehicle is calculated. Calculating the difference between the actual turning angle of the autonomous traveling carriage and the turning command angle by the difference in distance to each plate of the measuring jig measured before and after the turning, Calculate the odometry parameter at the time of turning, or calculate the difference between the actual turning angle and the turning command angle of the autonomous traveling vehicle by the difference in distance and angle to the flat plate measuring jig measured before and after turning. By calculating the odometry parameter during turning of the autonomous traveling vehicle, it is possible to easily adjust the odometry (wheel distance meter) parameter during turning.

  And in the adjustment method of the odometry parameter using the imaging device, the autonomous traveling vehicle needs to be charged periodically by installing the marker at the charging station of the autonomous traveling vehicle. Adjustment of odometry parameters can be performed, and an autonomous traveling vehicle that can always autonomously travel autonomously can be provided.

  Further, the setting of the autonomous traveling carriage at the start position or the goal position is performed by bringing the wheels into contact with a position shifted from the starting position or the goal position by a wheel radius of the autonomous traveling carriage. By using the positioning jig, manual installation can be facilitated and installation at the start position can be performed more accurately.

  Further, in the above odometry parameter adjustment method, the correction value for the error of each wheel was derived as an average value, but the movement distance of the autonomous traveling vehicle measured after the specific distance straight traveling, the start position, By calculating the diameter ratio of each wheel of the autonomous traveling vehicle from the inclination of the autonomous traveling vehicle with respect to the straight line connecting the goal positions, and adjusting the odometry (wheel distance meter) parameter of the autonomous traveling vehicle, each wheel It is also possible to adjust the odometry (wheel distance meter) parameter so as to individually correct the vehicle, and autonomous driving with higher accuracy becomes possible.

  As described above, according to the present invention, it is possible to automate and accurately adjust odometry parameters with a simple system configuration, reduce human intervention as much as possible, reduce the cost of autonomous traveling carts, and process up to product shipment. Shortening can be achieved.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, but are merely illustrative examples. Not too much.

  First, the odometry (wheel distance meter) parameter adjustment of the autonomous traveling vehicle will be described. In the odometry (wheel distance meter) parameter adjustment of the autonomous traveling vehicle, first, an error in actual operation with respect to the translation and turning command values is measured, and the parameter adjustment amount is calculated from the translation and turning errors. Then, the adjustment amount of this parameter is written in the parameter, and the error of actual operation with respect to the adjusted translation and turning command values is measured. Recheck the tread and repeat this adjustment.

  For example, wheel parameters in an autonomous traveling vehicle such as a robot are a wheel diameter and a tread (wheel interval). The straight wheel performance is affected by the wheel diameter, and the turning performance is affected by the wheel diameter and the tread. First, an error (in the x direction) in the case of translation (forward / backward) indicating straight running performance, the translation distance and the wheel diameter have the following relationship.

    Translation distance (mm) = π × rotational speed of wheels (average of left and right) × wheel diameter ……… (1)

  In addition, the turn on the spot has the following relationship due to the influence of the wheel diameter and the tread.

    Turning angle (rad) = π × difference in rotational speed of left and right wheels × wheel diameter / tread (2)

Therefore, the adjustment amount of the wheel diameter parameter is, first, the movement distance of the autonomous traveling carriage is L (odometry value), the adjustment amount of the wheel diameter parameter is kw, and the deviation between the command movement distance and the actual movement distance measurement value is Δx. Then
kw = | Δx | / L ………………………………………………… (3)
When the actual travel distance measurement value is longer than the command travel distance (Δx> 0), if the wheel diameter parameter is (WHEEL),
WHEEL = (1-kw) × WHEEL (4)
If the actual measured travel distance is shorter than the command travel distance (Δx <0),
WHEEL = (1 + kw) × WHEEL …………………… (5)
It becomes.

Next, the adjustment amount of the tread parameter is that the turning angle of the autonomous traveling vehicle is Θ (odometry value), the adjustment amount of the tread parameter is kt, and the deviation between the command turning amount and the actual turning amount actual measurement value is Δθ.
kw / kt = | Δθ | / Θ ………………………………………… (6)
When the actual turning amount actual measurement value is larger than the command turning amount (Δθ> 0), if the tread parameter is (TREAD),
TREAD = (1-kw / kt) × TREAD (7)
When the actual turning amount actual measurement value is smaller than the command turning amount (Δθ <0),
TREAD = (1 + kw / kt) × TREAD (8)
It becomes.

  That is, the odometry (wheel distance meter) parameter adjustment of the autonomous traveling vehicle is performed by calculating the values of kw and kt and adjusting the values of WHEEL and TREAD as described above.

  FIG. 1 is a diagram for explaining an embodiment 1 of an odometry (wheel distance meter) parameter adjusting method for an autonomous traveling carriage according to the present invention. FIG. 1A shows a start position and a goal position for measuring a straight traveling error. The figure shown, (B) is a figure for demonstrating the method to set an autonomous traveling vehicle to a start position with the positioning jig of an autonomous traveling vehicle, (C) is imaged by using a marker as a template before measuring a straight-running error. FIG. 2 is a diagram showing a positional relationship between the marker and the autonomous traveling vehicle in FIG. 2, FIG. 2 is a diagram showing an autonomous traveling robot as an example of the autonomous traveling vehicle, and a charging station having a marker installed at the goal position, FIG. Is a diagram showing a method of imaging a marker at the goal position, (B) is a diagram showing an example of a marker placed at the goal position, and FIG. 4 (A) is an autonomous traveling vehicle and marker at the time of turning error measurement. Figure, FIG. 5B is a diagram showing a positioning jig for installing the autonomous traveling vehicle at the position at the time of turning error measurement, FIG. 5C is a diagram showing the autonomous traveling vehicle after turning, and FIG. 5 is a flow for calculating a straight traveling error. FIG. 6 and FIG. 6 are flowcharts for calculating a turning error. In the following description, the case of a robot as an autonomous traveling vehicle will be described as an example. However, as described above, a beagle (AGV) that autonomously travels in a vehicle form as an autonomous traveling vehicle may be used. As long as the traveling wheels are at least two or more, a plurality of wheels may be provided.

  In the first embodiment of the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling vehicle according to the present invention, FIG. 2 includes an autonomous traveling robot 1 as an example of the autonomous traveling vehicle and a plurality of markers 20 installed at the goal position. As shown in the charging station 2, the robot 1 as an autonomous traveling carriage has a straight traveling error and a turning error by imaging the marker 20 with, for example, the imaging device 10 having a fish-eye lens or a wide-angle lens provided on the top of the head. Calculate and adjust odometry parameters. The imaging device for the autonomous traveling vehicle may be provided at any position as long as it can image the marker 20 as well as the top of the head as described above. The marker 20 also measures the distance from the autonomous traveling vehicle as described later. Any form can be used as long as it is possible.

  The actual odometry (wheel distance meter) parameter adjustment is performed according to the flowchart shown in FIG. 5 in the case of a straight running error, and in accordance with the flowchart shown in FIG. 6 in the case of a turning error.

  First, in the case of measuring a straight running error, first, in step S10 “positioning the carriage with the positioning jig at the origin” in the flow chart of FIG. 5, the robot 1 is moved as shown in FIG. For example, a goal position 31 (x, y = 50,000 mm, 0) away from the start position 30 (x, y = 0, 0) by a specific distance of, for example, 5 m is set to face the specific distance (5 m). Instruct and drive.

  At this time, when the robot 1 is installed at the start position 30, for example, when the traveling wheels of the robot 1 are two wheels, such as 11 and 12, the wheels 11 and 12 are placed at positions shifted from the start position 30 by the radius of the two wheels. If the positioning jig 3 having the portions 32 and 33 held in contact with each other is used, installation can be performed easily and accurately. In order to make the installation of the robot 1 at the start position 30 more accurate, a reference post (having a height lower than the wheel radius) 35 for bringing the positioning portion 34 of the positioning jig 3 into contact with the start position 30 is provided. It may be provided. When the positioning of the robot 1 is completed at the start position, the positioning jig 3 is slid to the rear of the robot 1 and removed as shown by the one-dot chain line.

  Furthermore, before actually running the robot 1 and measuring the straight running error, the robot 1 is set at the goal position 31 using the positioning jig 3 of the robot 1 as shown in FIG. Then, the marker 20 is imaged by the imaging device 10 at the top of the head, and the marker captured image when the normal goal position is reached is registered in advance in a storage means (not shown) as a template.

  FIG. 3 is a diagram showing an imaging method (A) of the marker 20 and an example (B) of the marker 20 installed at the goal position. As described with reference to FIG. It is provided at a position where it can be imaged by an imaging device provided with a fisheye lens or a wide angle lens. Then, the robot 1 is installed at the goal position 31 by the positioning jig 3 described above, and after traveling by a control means (not shown) from the interval between the plurality of markers 20 on the captured image and the distance to the marker of the robot 1. The distance between the marker 20 and the robot 1 can be calculated.

  Referring to FIG. 1 again, when the robot 1 is ready for the measurement of the straight running error of the robot 1, the robot 1 is set at the start position 30 by the positioning jig 3 as described above. In step S11 in the flow chart shown, for example, the above-mentioned straight-ahead instruction for a specific distance of 5 m is given and the vehicle travels straight. When the robot stops at the position where the robot has advanced 5 m in step S12, the marker 20 is imaged in the same manner as the statement made in FIG.

  In step S13, the distance between the robot 1 and the marker 20 is calculated from the imaged result and the previously captured template image, and the translation error at the stop position of the robot 1 is calculated. That is, the distance between the robot 1 and the marker 20 is calculated from the distance between the marker 20 and the robot 1 at the time of capturing the template image and the interval between the plurality of markers 20 on the template image as described above. By measuring the interval of 20, the distance between the marker 20 and the robot 1 can be calculated.

  Then, a difference Δx between the distance between the robot 1 and the marker 20 calculated from each captured image is calculated, thereby adjusting the odometry parameters of the robot 1 using the equations (3), (4), and (5). .

  When measuring a turning error, first, in step S20 “positioning the carriage with the positioning jig at the origin with the positioning jig” in the flowchart of FIG. 6, the robot 1 is moved to the goal position 31 as shown in FIG. 4B is placed with the marker 20 facing the same using the positioning jig 3 shown in FIG. 4B (the same as the positioning jig described in FIG. 1B), and the head is imaged in the next step S21. The marker 20 is imaged by the apparatus 10, and the marker image before turning is registered in advance in a storage means (not shown) as a template.

  When the robot 1 is ready for measuring the turning error of the robot 1 in this way, a turning command of 360 °, for example, is given to the robot 1 in step S22. An amount of rotation command is given to turn, and after the turn, the marker 20 is imaged again in step S23.

  In step S24, a deviation between the captured image and the template image registered in the storage means (not shown), that is, a turning error of the stop position is calculated. Based on the calculated value, (6), ( 7) The control means (not shown) adjusts the odometry parameter of the robot 1 using the equations (8). Now, at the end of this turn, if the robot 1 stops in a direction deviated by an angle Δθ from the normal direction as shown in FIG. 4C, for example, it deviates from the template image by this Δθ. An image is taken. Therefore, how many deviations of the image of the marker 20 correspond to is calculated from the distance between the marker 20 and the robot 1 at the time of capturing the template image and the interval between the plurality of markers 20 on the template image. The calculated angle Δθ is used to adjust the odometry parameter of the robot 1.

  The above is the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle of the first embodiment. Next, according to FIGS. 7 to 13, the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle of the second embodiment. Will be explained.

  In the first embodiment of the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling vehicle according to the present invention described above, the imaging device 10 provided with, for example, a fisheye lens or a wide-angle lens provided on the top of the robot 1 as the autonomous traveling vehicle. Thus, the straight running error and the turning error are calculated by imaging the marker 20, and the odometry parameters are adjusted. In the second embodiment, infrared rays and the like provided in the robot 1 as the autonomous traveling carriage are used. The odometry parameters are adjusted using the front and side obstacle detection sensors used.

  FIGS. 7A and 7B are diagrams for explaining front and side obstacle detection sensors provided in a robot as an example of an autonomous traveling vehicle. FIG. 7A is a front view, and FIG. FIG. 8 is a view for explaining an embodiment 2 of the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling carriage according to the present invention, and (A) is a start position and a goal position for measuring a straight traveling error. FIG. 9B is a diagram for explaining a method of setting the autonomous traveling vehicle to the start position by the positioning jig of the autonomous traveling vehicle, and FIG. 9A is a diagram for measuring the straight traveling error of the autonomous traveling vehicle. The figure which showed the measurement state before driving | running | working, (B) is the figure which showed the measurement state after driving | running | working similarly, FIG. 10 (A) is the figure which showed the state before the turn at the time of the turning error measurement of an autonomous running vehicle (B) installs an autonomous traveling vehicle at the turning error measurement position (C) is a diagram showing an autonomous traveling vehicle after turning, FIG. 11 (A) is a diagram showing a measurement state before turning at the time of turning error measurement of the autonomous traveling vehicle, FIG. 12B is a flowchart showing a measurement state after turning, FIG. 12 is a flowchart for calculating a straight traveling error, and FIG. 13 is a flowchart for calculating a turning error.

  Normally, autonomous vehicles such as Beagle (AGV) that autonomously travels in the form of a robot or vehicle determine whether there is a traveling surface such as the floor or the ground before traveling, or whether there are no obstacles in the traveling area, Accordingly, when there is an obstacle at the destination or when there is no running surface, it is necessary to take an action such as detouring or getting over the obstacle depending on the case. In order to detect these running surface and obstacle, FIG. As shown in FIG. 2, it is common practice to equip front and side obstacle detection sensors using infrared rays or the like.

In FIG. 7, 1 is a robot as an example of an autonomous traveling carriage, 11 and 12 are traveling wheels of the robot 1, and the robot 1 detects, for example, obstacles ahead, and a light emitting unit that outputs a light beam, for example. And a front obstacle detection sensor 41 having a light receiving portion for receiving the reflected light, and side obstacle detection sensors 42 ₁ and 42 ₂ provided on the left and right. Therefore, in the second embodiment of the present invention, the odometry parameters are adjusted using these front and side obstacle detection sensors. The front obstacle detection sensor 41 shown in FIG. 7 emits five light beams indicated as 1, No. 2, No. 3, No. 4, and No. 5 from the respective sensors. And a case of sensors 41 ₁ , 41 ₂ , 41 ₃ called wide angle sensors (WAS) having a light receiving portion for receiving the reflected light. Any type of sensor may be used as long as the distance can be measured in addition to the sensor.

  First, the straight running error is measured. First, “positioning the robot 1 to the origin position” in step S31 in the flowchart of FIG. 12 is performed. As shown in FIG. 8 (A), the robot 1 is moved to a start position 30 (x, y = 0, 0) by a goal position 31 (x, y = 50,000 mm, 0) It is installed with the direction facing, and is performed using the positioning jig 3 described with reference to FIG.

  Next, as shown in FIG. 8 at step S32 in FIG. Install so that. Then, before actually running the robot 1 and measuring the straight running error, in step S33 of FIG. 12, the front and side obstacle detection sensors 41 and 42 shown in FIG. The distance to each flat plate constituting the sensor detection value acquisition jig 40 as a tool is measured and stored in a storage means (not shown).

FIG. 9A is a diagram showing an example of the measurement result. For example, the flat plates 40 ₁ and 40 ₂ on the side of the robot 1 in the sensor detection value acquisition jig 40 are positioned at a position of 465 mm from the start position 30. There is the front of the flat plate 40 ₃ from the start position 30 to the position of 490 mm, further, the front of the plate detected by the 41 _second sensors of the forward obstacle sensor _{41 1,} 41 2, 41 ₃ provided in the robot 1 40 ₃ distance to the 300 mm _(Lr 1), also located in the side obstacle detection sensor 42 ₁ on the right (xr, yr), side obstacle detection sensor 42 _{and second} left there in (xl, yl) In this case, the distances to the side flat plates 40 ₁ and 40 ₂ detected by the respective sensors are 350 mm (Lr ₂ and Lr ₃ ), respectively. When the measurement before traveling is completed in this way, the positioning jig 3 is slid to the rear of the robot 1 and removed in step S34 of FIG.

Then, in step S35 of FIG. 12, the sensor detection value acquisition jig 40, which is the measurement jig described above, is installed at the start position 30 at the stop position (goal position) 31 of FIG. Install in. Next, in step S36 of FIG. 12, the robot 1 is made to travel straight by issuing, for example, the above-mentioned straight distance movement command of 5 m, and in step S37, as shown in FIG. Then, the distance to the side flat plates 40 ₁ , 40 ₂ and the traveling direction flat plate 40 _{3 in} the sensor detection value acquisition jig 40 is measured in the same manner as at the start position 30.

An example of this measurement result is shown in FIG. In this example, the robot 1 is passed the goal position 31, has stopped further by the flat plate 40 ₁ side, which is the traveling direction left of the robot 1 in the sensor detection value acquisition jig 40.

Now, mounted at an angle of θwas respect to the front obstacle detecting sensor 41 ₂ floor surface at this time, Lt ₁ distance detection result up to the front of the flat plate 40 _3, also side obstacle detection sensor ₄₂ 2, 42 flat ₄₀ 1 side by _1, 40 ₂ until the distance detection result of each _Lt 2, Lt _3, the distance by the distance that the robot 1 has gone too far forward [Delta] x, the flat plate 40 ₁ side [Delta] y, with respect to the x-direction Assuming that the amount of posture deviation is Δθ, the posture deviation Δθ is
Δθ = cos ⁻¹ {(Lr ₂ + | yl | + Lr ₃ + | yr |) /
(Lt ₂ + | yl | + Lt ₃ + | yr |)} (9)
It becomes. Further, the deviation Δx in the traveling direction is
Δx = Lr ₁ × cos θwas−Lt ₁ × cos θwas × cos Δθ (10)
The lateral deviation Δy is
Δy = Lr ₂ −Lt ₂ × cos Δθ (11)
It becomes.

Therefore, as described above, if the wheel diameter parameter adjustment amount kw and the wheel diameter parameter are WHEEL,
kw = | Δx | / L ………………………………………………… (3)
When the actual measured travel distance is longer than the command travel distance (Δx> 0),
WHEEL = (1-kw) × WHEEL (4)
If the actual measured travel distance is shorter than the command travel distance (Δx <0),
WHEEL = (1 + kw) × WHEEL …………………… (5)
Thus, these calculations are performed by control means not shown.

  Next, when measuring a turning error, first, “positioning the robot to the origin position” in step S41 in the flowchart of FIG. 13 is performed. As shown in FIG. 10A, the robot 1 is set to the start position 30 (x, y = 0, 0), and the positioning jig 3 described in FIG. 1 (see FIG. 10B) is used. Use it.

  Next, in step S42 in FIG. 13, the sensor detection value acquisition jig indicated by 40 in FIG. 10 is installed so that each flat plate is a predetermined distance from the start position 30. Then, before actually turning the robot 1, in step S43 of FIG. 13, the sensor detection value is obtained as a measurement jig using the front and side obstacle detection sensors 41 and 42 shown in FIG. The distance to each flat plate constituting the jig 40 is measured.

FIG. 11A is a diagram showing an example of the measurement result. As in the case of FIG. 9A, for example, the flat plate 40 ₁ , 40 on the side of the robot 1 in the sensor detection value acquisition jig 40. ₂ is at a position of 465 mm from the start position 30, and the front flat plate 40 ₃ is at a position of 490 mm from the start position 30. Further, of the front obstacle detection sensors 41 ₁ , 41 ₂ , 41 ₃ provided in the robot 1 front at 41 _second sensor distance Lr ₁ up to the front of the flat plate 40 ₃ detected by the sensor (center No.3 sensors of that shown in FIG. 7 to 41 ₂₎ is 300 mm, the left and right in the sensor of the 41 ₂ sensor distance _Lr up front of the flat plate 40 ₃ detected by the (No.1 of right and left shown in FIG. 7 to 41 _2, sensors No.5) 4, Lr _5, also right side obstacle detection sensor 42 ₁ (xr Located yr), side obstacle detection sensor 42 _{and second} left (xl, be in yl), flat ₄₀ 1 lateral detected by each sensor, 40 a distance of up to _2, each 350 mm _(Lr 2, Lr ₃ ) is shown. When the measurement before running is completed in this way, in step S44 of FIG. 13, the positioning jig 3 is slid to the rear of the robot 1 and removed in the same manner as described above, and then a turning command of 360 °, for example, is given to the robot 1 to turn. .

Then, in step S45 of FIG. 12, the flat plates 40 ₁ , 40 ₂ on the side of the jig 40 for sensor detection value acquisition and the flat plates on the front side are stopped as shown in FIG. The distance to 40 ₃ is measured with each sensor.

An example of the measurement result is shown in FIG. In this example, the robot 1 has a diameter difference of the left and right wheels, are stopped by the flat plate 40 ₁ side, which is from the original position 30 to the traveling direction left front of the flat plate 40 in _three directions and the robot 1.

Now, the front obstacle detection sensor 41 ₂ on the front Lt 1 the distance to the front of the flat plate 40 ₃ detected by the sensor (Figure 7 in the middle of the No.3 sensors shown in 41 _2), the right lateral fault object detecting sensor 42 ₁ is (xr, yr) is in the lateral obstacle sensor 42 ₂ on the left side (xl, yl) be in until the flat plate ₄₀ 1, 40 ₂ lateral detected by each sensor 2 distance each _Lt, Lt _3, the front obstacle detection sensor 41 right and left sensors in the _two front flat plate 40 ₃ detected by the (right and left No.1 shown in FIG. 7 to 41 _2, sensors No.5) Lt ₄ and Lt ₅ , the distance that the robot 1 is close to the flat plate 40 ₃ direction is Δx, the distance by the flat plate 40 ₁ side is Δy, and the posture deviation amount that the posture is inclined with respect to the x direction is Δθ. To do.

In calculating the amount of deviation of the posture, first, the positive / negative judgment of the posture is performed. At this time, the front obstacle detection sensor 41 right and left of the sensor at _second distance _Lt 4 up to the front of the flat plate 40 ₃ detected (Fig. 7 to the right and left No.1 shown in 41 _2, sensors No.5), Lt _{No. 5} is equal if the left and right sensors are attached at the correct positions, but the detection value before turning is corrected in consideration of the case where a difference in measurement distance occurs due to attachment errors or the like. That is, when the difference between the distances Lt ₄ and Lt ₅ is ΔL,
ΔL = Lt ₄ −Lt ₅
When ΔL> 0 Lt ₅ = Lt ₅ + ΔL
When ΔL <0 Lt ₄ = Lt ₄ + ΔL
And the sign of Δθ is
Δθ> 0 when Lt ₄ −Lt ₅ > 0
Lt ₄ -Lt ₅ <Δθ <0 when 0
It becomes.

Therefore, the posture deviation Δθ is Δθ = cos ⁻¹ {(Lr ₂ + | yl | + Lr ₃ + | yr |) /
(Lt ₂ + | yl | + Lt ₃ + | yr |)} (12)
It becomes.

Then, the adjustment amount of the tread parameter is that the turning angle of the autonomous traveling carriage is Θ (odometry value), the adjustment amount of the tread parameter is kt, and the deviation between the command turning amount and the actual turning amount actual measurement value is Δθ ′,
Δθ ′ = 360 ° −Θ ……………………………………………………… (13)
kw / kt = | Δθ−Δθ ′ | / …………………………………… (14)
When the actual turning amount actual value is larger than the command turning amount (Δθ−Δθ ′> 0), when the tread parameter is (TREAD),
TREAD = (kt / (kt + kw)) × TREAD
= (Θ / (Θ + (Δθ−Δθ ′))) × TREAD (15)
When the actual turning amount actual measurement value is smaller than the command turning amount (Δθ−Δθ ′ <0),
TREAD = (kt / (kt−kw)) × TREAD
= (Θ / (Θ− (Δθ−Δθ ′))) × TREAD (16)
It becomes.

  That is, as described above, the odometry (wheel distance meter) parameter adjustment of the autonomous traveling vehicle is performed by calculating the values of kw and kt by the control means (not shown) and thereby adjusting the values of WHEEL and TREAD. That is why.

  The above is the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling vehicle of the second embodiment. Next, according to FIGS. 14 to 17, the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling vehicle of the third embodiment. Will be explained.

In the second embodiment of the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling vehicle according to the present invention described above, the front and side obstacles using infrared rays and the like provided in the robot 1 as the autonomous traveling vehicle are described. Although the detection sensor and the sensor detection value acquisition jig in which a flat plate is formed in a substantially U shape are used as the measurement jig, in the third embodiment, for example, the wide angle sensor ( WAS) In the front obstacle detection sensor composed of 41 ₁ , 41 ₂ , 41 ₃ , only at least two beams having an angular difference capable of measuring the inclination of the autonomous traveling carriage are used, and a flat plate sensor is used as a measurement jig The odometry parameter is adjusted using a detection value acquisition jig.

  FIG. 14 is a diagram for explaining a third embodiment of the odometry (wheel distance meter) parameter adjusting method of the autonomous traveling carriage according to the present invention, and (A) shows the start position and goal position for measuring the straight traveling error. FIG. 15B is a diagram for explaining the influence of the posture error on the adjustment of the wheel diameter parameter, and FIG. 15 is a diagram for explaining a method for measuring the posture of the autonomous traveling carriage after straight traveling. 16 is a flowchart for calculating a straight traveling error, and FIG. 17 is a flowchart for calculating a turning error.

  First, in the case of measuring a straight traveling error, first, “install the robot at the measurement start position and reset the position and orientation to 0” in step S51 in the flowchart of FIG. As shown in FIG. 14 (A), the robot 1 is moved to the start position 30 (x, y = 0, 0) by a goal position 31 (x, y = 50,000 mm, 0) Orientation is set and a flat sensor detection value acquisition jig 50 is installed as a dedicated measurement jig.

Next, a beam of the front of the obstacle detection sensor 41, for example, the center of the 41 ₂ No.3 shown in FIG 7 at step S52 in FIG. 16, left and right 41 ₁ No.1 beams 41 ₃ No. 5 beam or the like is used to measure the distance to the flat plate constituting the flat plate sensor detection value acquisition jig 50 and store it in a storage means (not shown).

Now, parameters such as the positions of the center 41 ₂ , the right 41 ₁ , and the left 41 _{3 in} the front obstacle detection sensor 41 are defined as follows.

(1) Center front obstacle detection sensor 41 ₂
Mounting position: (x ₁ , y ₁ )
Mounting angle (vertical direction): θ ₀ [deg]
Mounting angle (horizontal direction): θ ₁ [deg]
(2) Left front obstacle detection sensor 41 ₃
Mounting position: (x ₂ , y ₂ )
Mounting angle (vertical direction): θ ₀ [deg]
Mounting angle (horizontal direction): θ ₂ [deg]
(3) Right front obstacle detection sensor 41 ₁
Mounting position: (x ₃ , y ₃ )
Mounting angle (vertical direction): θ ₀ [deg]
Mounting angle (horizontal direction): θ ₃ [deg]

  And each measurement result is defined as follows.

(1) the center of the front obstacle detection sensor 41 _{and second} measurement results: _L 1 [mm]
(2) Measurement result of left front obstacle detection sensor 41 ₃ : L ₂ [mm]
(3) Measurement result of right front obstacle detection sensor 41 ₁ : L ₃ [mm]

Next, in step S53 of FIG. 16, the initial position and orientation are calculated from these measurement results. As shown in FIG. 15, this calculation is actually performed as shown in FIG. 15A, in which the robot 1 is tilted with respect to the flat sensor detection value acquisition jig 50. The inclination is calculated assuming that the tool 50 is inclined with respect to the robot 1 as shown in FIG. At this time, since the attachment position and orientation of the sensor are known with the center coordinates of the robot 1 as the origin, the coordinates (x _L , y _L ) and (x _R , y _R ) are obtained from the measurement results L ₂ and L _3. The flat sensor detection value acquisition jig 50 and the inclination θ of the robot 1 can be calculated from the straight line passing through these two points.

Therefore, the initial posture of the inclination θ _{S of the} robot 1 with respect to the flat sensor detection value acquisition jig 50 is calculated using the following equation.

このとき、前方障害物検知用センサ４１の取り付け位置から平板のセンサ検出値取得用治具５０までの水平距離Ｌ_Ｓは次の式で求められる。 x _L1 = x ₂ + L ₂ cos θ ₀ cos θ ₂ (17)
y _L1 = y ₂ + L ₂ cos θ ₀ cos θ ₂ (18)
x _R1 = x ₃ + L ₃ cos θ ₀ cos θ ₃ ……………………………… (19)
y _R1 = y ₃ + L ₃ cos θ ₀ cos θ ₃ ……………………………… (20)

At this time, the horizontal distance L _S from the attachment position of the front obstacle detection sensor 41 to the flat sensor detection value acquisition jig 50 is obtained by the following equation.

L _S = L ₁ cos θ _S (θ _S ≦ 3.0 deg → cos θ _S = 1) (22)

  Then, a flat sensor detection value acquisition jig 50 as a measurement jig is similarly installed at a measurement distance of, for example, 5 m from the start position 30. When the installation is completed, in step S54 of FIG. 16, the robot 1 is caused to issue, for example, a linear movement command with a specific distance of 5 m as described above, and the robot 1 is stopped. The flat sensor detection value acquisition jig 50 is used for measurement.

  And each measurement result is defined as follows.

(1) the center of the front obstacle detection sensor 41 _{and second} measurement results: _L 4 [mm]
(2) Measurement result of left front obstacle detection sensor 41 ₃ : L ₅ [mm]
(3) Measurement result of right front obstacle detection sensor 41 ₁ : L ₆ [mm]

  At this time, when the robot 1 is moved over a long distance, the robot 1 moves as shown in FIG. 14B when tilted with respect to an ideal long-distance movement path orthogonal to the sensor detection value acquisition jig 50. Thus, a cosine error occurs and becomes a problem. However, if the initial posture error is within 3.6 degrees, for example, the error when moving 50000 mm is within 10 mm, which is an allowable range for the target positioning accuracy. Further, since the front obstacle detection sensor 41 naturally has variations, it is necessary to calculate the posture error by an algorithm that takes this into consideration.

  Next, in step S56 of FIG. 16, the positioning error with respect to the movement command is calculated from the measurement result of the front obstacle detection sensor 41, and the adjustment amount of the wheel diameter parameter is calculated. Therefore, first, the inclination of the robot 1 with respect to the flat sensor detection value acquisition jig 50 installed at the stop position is measured, and the attitude of the stop position is calculated. The calculation formula is as follows.

このとき、前方障害物検知用センサ４１の取り付け位置から平板のセンサ検出値取得用治具５０までの水平距離Ｌ_Ｆは次の式で求められる。 x _L2 = x ₂ + L ₅ cos θ ₀ cos θ ₂ ………………………………… (23)
y _L2 = y ₂ + L ₅ cos θ ₀ cos θ ₂ ………………………………… (24)
x _R2 = x ₃ + L ₆ cos θ ₀ cos θ ₃ ………………………………… (25)
y _R2 = y ₃ + L ₆ cos θ ₀ cos θ ₃ ………………………………… (26)

At this time, the horizontal distance L _F from the attachment position of the front obstacle detection sensor 41 to the flat sensor detection value acquisition jig 50 is obtained by the following equation.

L _F = L ₄ cos θ _F (θ _F ≦ 3.0 deg → cos θ _F = 1) (28)

Further, a movement error ΔL ₁ taking into account the posture deviation after the long-distance straight-ahead operation is calculated by the following equation.

さらに、この移動誤差ΔＬ_１を使って、オドメトリ量Ｌ_ｏｄと動作指令値との誤差ΔＬ_２などの車輪径パラメータの補正量を算出する。 Δθ = | θ _F −θ _S |

Furthermore, with this movement error [Delta] L _1, it calculates a correction amount of the wheel diameter parameters such as an error [Delta] L ₂ between the operational command value odometry amount L _od.

ΔL _od = L _od −5000 [mm] ……………………………… (30)

となり、指令移動距離より実際の移動距離実測値が長い（ΔＬ＞０）場合、
ＷＨＥＥＬ＝（１＋ｋｗ）×ＷＨＥＥＬ ……………………………………（３３）
指令移動距離より実際の移動距離実測値が短い（ΔＬ＜０）場合、
ＷＨＥＥＬ＝（１−ｋｗ）×ＷＨＥＥＬ ……………………………………（３４）
となる。 Therefore, as described above, if the wheel diameter parameter adjustment amount kw and the wheel diameter parameter are WHEEL,

When the actual travel distance measurement value is longer than the command travel distance (ΔL> 0),
WHEEL = (1 + kw) × WHEEL ……………………………… (33)
If the actual measured travel distance is shorter than the command travel distance (ΔL <0),
WHEEL = (1-kw) × WHEEL ……………………………… (34)
It becomes.

  When the parameter adjustment amount is calculated in this way, the process proceeds to step S57 in FIG. 16 and after adjusting the parameter, the movement test is performed again, and if the positioning error is within 10 mm for the 5000 mm linear movement, the test is accepted. If it is 10 mm or more, the wheel diameter parameter is returned to the default, and adjustment is performed by performing a test from the beginning. If this value is within 10 mm, the process proceeds to the tread parameter adjustment test in the next step S58.

In the tread parameter adjustment, as described above, “install the robot at the measurement start position and reset the position and orientation to 0” is performed in step S61 of the flowchart shown in FIG. Then, as in the Step S62, using the forward obstacle detecting sensor 41 shown in FIG. 7, for example, left and right 41 ₁ No.1 beam, and beam 41 ₃ No.5, flat plate The distance to the flat plate constituting the sensor detection value acquisition jig 50 is measured.

At this time, parameters such as the positions of the right 41 ₁ and the left 41 ₃ in the front obstacle detection sensor 41 are defined as described above.

(4) Left front obstacle detection sensor 41 ₃
Mounting position: (x ₂ , y ₂ )
Mounting angle (vertical direction): θ ₀ [deg]
Mounting angle (horizontal direction): θ ₂ [deg]
(5) Right front obstacle detection sensor 41 ₁
Mounting position: (x ₃ , y ₃ )
Mounting angle (vertical direction): θ ₀ [deg]
Mounting angle (horizontal direction): θ ₃ [deg]

  In addition, the measurement result at each initial position is defined as follows.

(4) Measurement result of left front obstacle detection sensor 41 ₃ : L ₂ [mm]
(5) Measurement result of right front obstacle detection sensor 41 ₁ : L ₃ [mm]

Next, in step S63 of FIG. 17, the inclination θ _{S of the} robot 1 with respect to the flat plate sensor detection value acquisition jig 50 is calculated from these measurement results using the following equation.

x _L1 = x ₂ + L ₂ cos θ ₀ cos θ ₂ (17)
y _L1 = y ₂ + L ₂ cos θ ₀ cos θ ₂ (18)
x _R1 = x ₃ + L ₃ cos θ ₀ cos θ ₃ ……………………………… (19)
y _R1 = y ₃ + L ₃ cos θ ₀ cos θ ₃ ……………………………… (20)

  Next, in step S64 of FIG. 17, a 360 degree rotation command is given to the robot 1, and in step S65, measurement is performed with the flat plate sensor detection value acquisition jig 50. Each measurement result is defined as follows as described above.

(4) Left front obstacle detection sensor 41 ₃ : L ₅ [mm]
(5) Right front obstacle detection sensor 41 ₁ : L ₆ [mm]

  Next, in step S66 of FIG. 17, an attitude error with respect to the rotation command is calculated from the measurement result of the forward obstacle detection sensor 41, and an adjustment amount of the tread parameter is calculated. Therefore, as described above, first, the inclination of the robot 1 with respect to the flat sensor detection value acquisition jig 50 installed at the stop position is measured, and the attitude of the stop position is calculated. The calculation formula is as follows.

x _L2 = x ₂ + L ₅ cos θ ₀ cos θ ₂ ………………………………… (23)
y _L2 = y ₂ + L ₅ cos θ ₀ cos θ ₂ ………………………………… (24)
x _R2 = x ₃ + L ₆ cos θ ₀ cos θ ₃ ………………………………… (25)
y _R2 = y ₃ + L ₆ cos θ ₀ cos θ ₃ ………………………………… (26)

  Then, an error from the actual turning amount with respect to the 360-degree turning operation command is calculated based on the following equation.

Δθ = θ _F −θ _S ………………………………………………………… (35)

As described above, the adjustment amount of the tread parameter is that the turning angle of the autonomous traveling vehicle is Θ (odometry value), the adjustment amount of the tread parameter is kt, and the difference between the command turning amount and the actual turning amount measured value is Δθ ′. Then,
Δθ ′ = 360 ° −Θ ……………………………………………………… (13)
kw / kt = | Δθ−Δθ ′ | / …………………………………… (14)
When the actual turning amount actual value is larger than the command turning amount (Δθ−Δθ ′> 0), when the tread parameter is (TREAD),
TREAD = (kt / (kt + kw)) × TREAD
= (Θ / (Θ + (Δθ−Δθ ′))) × TREAD (15)
When the actual turning amount actual measurement value is smaller than the command turning amount (Δθ−Δθ ′ <0),
TREAD = (kt / (kt−kw)) × TREAD
= (Θ / (Θ− (Δθ−Δθ ′))) × TREAD (16)
It becomes.

  When the parameter adjustment amount is calculated in this way, the process proceeds to step S67 in FIG. 17, and after adjusting the parameter, the rotation test is performed again. If it is 3 degrees or more, the tread parameter is returned to the default, and the test is adjusted from the beginning to adjust. And if this value is within 3 degrees, it ends as a pass.

  The above is the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle of the third embodiment. Next, the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle of the fourth embodiment will be described with reference to FIG. .

  In the first to third embodiments of the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle according to the present invention described above, an obstacle using an imaging device or infrared rays provided in the robot 1 as the autonomous traveling vehicle. Although the object detection sensor is used, in the fourth embodiment, the odometry of the autonomous traveling vehicle is detected by the position / orientation detection sensor that detects the position and orientation of the autonomous traveling vehicle 1 provided in the traveling space of the autonomous traveling vehicle. Wheel distance meter) parameter is adjusted.

  FIG. 18 is a diagram for explaining an embodiment 4 of the odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to the present invention. The robot 1 as an autonomous traveling vehicle combines reflectors at substantially right angles. The posture measuring jig 60 is mounted and installed at the start position 30 so that one side of the posture measuring jig 60 is parallel to the goal position 31 direction. Further, the position / posture detection sensor for detecting the position and posture of the robot 1 is located on the straight line connecting the start position 30 and the goal position 31 to the position / posture detection sensor mounting member 70 in the vicinity of the goal position. Sensors 71 and 72 for irradiating infrared rays and the like at the same angle θ, and sensors 72 for irradiating infrared rays and the like in the direction perpendicular to the straight line connecting the start position 30 and the goal position 31 from the goal position. One side of the position / posture detection sensor mounting member 70 is parallel to a straight line connecting the start position 30 and the goal position 31.

Now, the X-direction distance and Y-direction distance from the start position 30 of the reflection plate in the X and Y directions in the posture measuring jig 60 attached to the robot 1 are Dx and Dy, respectively, the travel command distance is L, The distances from the goal position 31 of the sensors 71, 72 and sensor 73 attached to the posture detection sensor attachment member 70 are respectively set to Ex, Ey, and the posture measurement from the sensors 71, 72 and sensor 73 at the position reached by the robot 1 by the travel command. The distance to the jig 60 is F ₁ , F ₂ , F ₃ , the coordinates at which the beam from each sensor has reached the attitude measurement jig 60, the sensor 71 is (x ₁ , y ₁ ), and the sensor 72 is ( x ₂ , y ₂ ), sensor 73 (x ₃ , y ₃ ), robot 1 coordinates (x _r , y _r ) at the position reached by robot 1 by the travel command, posture measurement The coordinates of the position at a right angle in the jig 60 for use are defined as (x _o , _yo ).

Further, when each side of the posture measuring jig 60 at the position where the robot 1 has reached by the travel command is regarded as a straight line, it is expressed by an equation, y = f ₁ (x), y = f ₂ (x ), The straight lines represented by the respective equations intersect at right angles, and y = f ₁ (x) passes through the coordinates (x ₁ , y ₁ ), (x ₂ , y ₂ ), and y = f ₂ (X) passes through the coordinates (x ₃ , y ₃ ).

  for that reason,

となる。

It becomes.

Accordingly, the posture at the position (x _r , y _r ) where the robot 1 has reached by the travel command, that is, the angle γ made by the robot 1 with respect to the straight line connecting the start position 30 and the goal position 31 is tan γ = (y ₂ −y _{1) / (x 2 -x 1} )
Therefore, γ = tan ⁻¹ {(y ₂ −y ₁ ) / (x ₂ −x ₁ )}
The coordinates (x _r , y _r ) of the robot 1 are

  Therefore, when measuring the straight running error as described above, the robot 1 is placed at the start position 30 in FIG. 18 and one side of the posture measuring jig 60 is parallel to a straight line connecting the start position 30 and the goal position 31. The sensor 71, 72, and sensor 73 attached to a position / orientation detection sensor attachment member 70 provided near the goal position 31 are provided with a travel command for a distance L, and are directed toward the goal position 31. By measuring the distance to the measuring jig 60, the actual travel distance of the robot 1 and the posture after the travel can be calculated.

  When measuring the turning error, the robot 1 is placed at the goal position 31 so that one side of the posture measuring jig 60 is parallel to a straight line connecting the start position 30 and the goal position 31 and is set at a predetermined angle. An actual turning angle of the robot 1 is measured by giving a turning command and measuring the distance to the posture measuring jig 60 with the sensors 71, 72 and sensor 73 attached to the position / posture detection sensor mounting member 70. Can be calculated.

  By doing in this way, even if it is an autonomous traveling trolley | bogie which does not have an imaging device or a sensor, an odometry (wheel distance meter) parameter can be adjusted easily.

  The wheel diameter parameter WHEEL calculated by the method described above derives a correction value for the error from the average value of the left and right wheels. However, when the robot 1 as the autonomous traveling carriage moves while tilting from the route with respect to the straight movement command, there is a difference in diameter between the left and right wheels, and the posture deviation Δθ calculated by the equation (9), Using the deviation Δx in the traveling direction calculated in 10) and the lateral deviation Δy calculated in Equation (11), the diameter difference between the left and right wheels can be calculated, and the left and right wheels can be individually corrected. It becomes.

FIG. 19 is a diagram for explaining a method of calculating the diameter difference of the traveling wheels from the traveling distance and the inclination of the posture after traveling of the autonomous traveling carriage 1. Now, of the traveling wheels of the autonomous traveling vehicle 1, for example, 11 is the larger wheel, the wheel diameter is φ _L , the travel distance is L _L , and 12 is the smaller wheel, the wheel diameter is φ _S. The distance is L _S , the distance between the two wheels (tread) is T, n is the number of wheel rotations, L is the distance actually moved with respect to the travel command (a value obtained by subtracting the error Δx from the travel command amount), and the autonomous traveling vehicle After the movement, the angle inclined with respect to the straight line connecting the start position and the goal position is Δθ, and the autonomous traveling vehicle 1 travels in an arc depending on the diameter difference between the traveling wheels 11 and 12. Is R.

Then, the radius R of the arc drawn by the autonomous vehicle 1 is
R = L / sinΔθ …………………………………………… (39)
Thus, the travel distances of the wheel 11 having a large diameter (wheel diameter φ _L ) and the wheel 12 having a small diameter (wheel diameter φ _S ) are L _L = 2π (R + T / 2) × Δθ / 360 = πnφ _L (40) )
L _S = 2π (R−T / 2) × Δθ / 360 = πnφ _S (41)
It becomes. Therefore, the ratio of each wheel diameter is
φ _L / φ _S = (R + T / 2) / (R−T / 2) (42)
It becomes.

  Therefore, by calculating the wheel diameter parameter WHEEL using this equation (39), the left and right wheels can be individually corrected as described above.

  In the above description, a method for adjusting an odometry parameter by traveling an autonomous traveling vehicle such as a robot for a predetermined distance has been described. However, in this method, a place of a certain size is required for traveling the autonomous traveling vehicle. Is required.

  Therefore, for example, a rotating roller or a rotating belt may be installed on the floor, and the autonomous traveling carriage may travel on the same to adjust the parameters in the same manner. In this case, a space where the autonomous traveling vehicle travels is unnecessary, and a sensor for measuring the distance from the autonomous traveling vehicle is also unnecessary.

  However, in this method, when a rotating belt is used or the rotating roller is driven in the same direction, the turning motion cannot be measured. However, for example, the driving wheel of the autonomous traveling carriage is sandwiched between two rotating rollers provided with an encoder. If the rotational speed of the drive wheel is detected, the turning motion can be measured.

  As described above, according to the present invention, it is possible to automate and accurately adjust odometry parameters with a simple system configuration, and to reduce the cost of an autonomous traveling vehicle by reducing human intervention as much as possible, as well as product shipment. Process shortening up to can be achieved.

  ADVANTAGE OF THE INVENTION According to this invention, the autonomous traveling vehicle which can drive | work correctly in the narrow space which has various obstructions, such as a room | chamber interior and a factory, can be provided at a low cost for a short period of time.

BRIEF DESCRIPTION OF THE DRAWINGS It is the figure for demonstrating Example 1 of the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle which becomes this invention, (A) is the figure which showed the start position and the goal position for a straight traveling error measurement, (B) is a figure for demonstrating the method of setting an autonomous traveling vehicle to a start position with the positioning jig of an autonomous traveling vehicle, (C) is a marker in the case of imaging a marker as a template before measuring a straight traveling error, It is the figure which showed the positional relationship of the autonomous traveling vehicle. It is the figure which showed the charging station which has the autonomous traveling robot as an example of an autonomous traveling cart, and the marker installed in a goal position. (A) is the figure which showed the method of imaging a marker in a goal position, (B) is the figure which showed an example of the marker installed in a goal position. (A) is a diagram showing an autonomous traveling vehicle and a marker at the time of turning error measurement, (B) is a diagram showing a positioning jig for installing the autonomous traveling vehicle at a position at the time of turning error measurement, (C) is a figure after turning It is the figure which showed the autonomous traveling vehicle. It is a flowchart in the case of calculating a straight running error. It is a flowchart in the case of calculating a turning error. It is a figure for demonstrating the sensor for a front and side obstruction detection provided in the robot as an example of an autonomous traveling trolley, (A) is a front view, (B) is XX direction sectional drawing. It is a figure for demonstrating Example 2 of the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle which becomes this invention, (A) is the figure which showed the start position and goal position for a straight traveling error measurement (B) is a figure for demonstrating the method of setting an autonomous traveling vehicle to a start position with the positioning jig of an autonomous traveling vehicle. (A) is the figure which showed the measurement state before the driving | running | working for the straight driving | running | working driving | running | working error measurement of an autonomous running cart, (B) is the figure which showed the measurement state after driving | running | working similarly. (A) is a view showing a state before turning at the time of turning error measurement of an autonomous traveling vehicle, (B) is a view showing a positioning jig for installing the autonomous traveling vehicle at a turning error measurement position, and (C) is turning. It is the figure which showed the rear autonomous traveling trolley. (A) is the figure which showed the measurement state before the turn at the time of the turning error measurement of an autonomous traveling trolley, (B) is the figure which showed the measurement state after the turn similarly. It is a flowchart in the case of calculating a straight running error. It is a flowchart in the case of calculating a turning error. It is a figure for demonstrating Example 3 of the odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle which becomes this invention, (A) is the figure which showed the start position and goal position for a straight traveling error measurement (B) is a figure for demonstrating the influence which attitude | position error has on adjustment of a wheel diameter parameter. It is a figure for demonstrating the measuring method of the attitude | position of the autonomous running cart after straight running. It is a flowchart in the case of calculating a straight running error. It is a flowchart in the case of calculating a turning error. It is a figure for demonstrating Example 4 of the odometry (wheel distance meter) parameter adjustment method of the autonomous running vehicle which becomes this invention. It is a figure for demonstrating the method of calculating the diameter difference of a driving | running | working wheel from the driving | running | working distance and inclination of an attitude | position after driving | running | working of an autonomous traveling vehicle.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Autonomous traveling robot 10 Imaging device 11, 12 Traveling wheel 2 of robot 1 Charging station 20 Marker 3 Positioning jig 30 Start position 31 Goal position 32, 33 Part 34 holding traveling wheel Positioning part 35 Reference post (height (Lower than wheel radius)

Claims (21)

In an autonomous traveling vehicle that travels with at least two traveling wheels each having an independent drive source and an imaging device, the travel distance is determined by the amount of rotation of the traveling wheel and the difference in the amount of rotation of each traveling wheel. Odometry function (movement distance measurement function) for estimating the moving direction, and an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating the moving distance and the moving direction A method for adjusting the odometry (wheel distance meter) parameter of an autonomous traveling vehicle,
After setting a marker that can be imaged by the imaging device in the vicinity of a goal position at a specific distance from the start position of the autonomous traveling carriage, and preparing an image of the marker previously captured by the imaging device at the goal position as a template The autonomously traveling carriage is installed at the start position in the goal direction and gives a travel instruction for the specific distance, or a turn command for one rotation is given at the goal position, and the marker imaged after running or after turning An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle, wherein the odometry (wheel distance meter) parameter is adjusted using the captured image and the template image.   The marker is composed of at least two markers, and the distance between the autonomous traveling vehicle and the marker after traveling using the image of the marker previously captured by the imaging device at the goal position and the distance to the marker of the autonomous traveling vehicle. The odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle according to claim 1, wherein the distance between the markers is determined such that the distance between the markers can be calculated.   Calculating a difference between the goal position and the actual arrival position of the autonomous traveling vehicle from the difference between the marker image captured after the traveling and the template image captured in advance, and calculating an odometry parameter when the autonomous traveling vehicle is traveling straight ahead. An odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to claim 1 or 2.   A turning angle error of the autonomous traveling vehicle is calculated from a difference between the marker image set at the goal position and imaged after the turn and the pre-captured template image, and an odometry parameter at the time of turning of the autonomous traveling vehicle is calculated. The odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to claim 1 or 2.   The odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to any one of claims 1 to 4, wherein the marker is installed at a charging station of the autonomous traveling vehicle. The amount of rotation of the traveling wheel and the amount of rotation of each traveling wheel in an autonomous traveling vehicle traveling with at least two traveling wheels having independent drive sources and front and side obstacle detection sensors. An odometry function (movement distance measurement function) for estimating a movement distance and a movement direction based on the difference, and an odometry determined by a wheel diameter and a wheel interval (tread) for estimating the movement distance and the movement direction. A method for adjusting an odometry (wheel distance meter) parameter of an autonomous traveling vehicle, which adjusts a (wheel distance meter) parameter,
A measuring jig having a substantially U-shaped flat plate is prepared and installed so that the distance from the starting position of the autonomous traveling carriage to each flat plate constituting the measuring jig is a predetermined distance. After setting the traveling carriage toward the goal position separated by a specific distance, after calculating the distance from the front and side obstacle detection sensors to each flat plate in the measurement jig in advance,
When measuring the turning angle error, the turning command for one rotation is directly used. When calculating the running error, the measuring jig is installed at the goal position to give the running command for the specific distance, and the front after turning or after running And the distance to each flat plate of the measurement jig by the side obstacle detection sensor is calculated, and the odometry (wheel distance meter) parameter of the autonomous traveling vehicle is calculated from the difference from the value measured before the turning or the traveling. An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle characterized by adjusting.   The difference between the actual position of the autonomous traveling carriage and the goal position is calculated based on the difference between the starting position and the distance to each plate of the measurement jig measured after traveling, and the autonomous traveling carriage travels straight. An odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to claim 6, wherein an hour odometry parameter is calculated.   The difference between the actual turning angle of the autonomous traveling vehicle and the turning command angle is calculated from the difference in distance to each flat plate of the measuring jig measured before and after the turning, and the odometry parameter when turning the autonomous traveling vehicle The odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle according to claim 6 or 7, wherein: An autonomous traveling vehicle that has at least two traveling wheels each having an independent drive source and an obstacle detection sensor that includes at least two beams having an angular difference capable of measuring the inclination of the autonomous traveling vehicle. The odometry function (movement distance measurement function) for estimating the movement distance and the movement direction based on the rotation amount of the traveling wheel and the difference between the rotation amounts of the respective traveling wheels, the movement distance and the movement direction An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimation,
A flat plate measuring jig capable of reflecting the beam in the obstacle detection sensor is prepared, and the autonomous traveling carriage is installed at a start position at a goal position separated by a specific distance and the flat plate measuring jig is measured. After calculating the autonomous traveling carriage posture angle from the distance to the flat plate measurement jig and the irradiation angle and detection distance of each beam by each beam in the obstacle detection sensor in advance,
When measuring the turning angle error, the turn command for one rotation is used as it is, and when calculating the running error, the flat plate measuring jig is installed at the goal position to give the running command for the specific distance. The autonomous traveling vehicle attitude angle is calculated from the distance to the flat plate measuring jig by the respective beams in the obstacle detection sensor, the irradiation angle of each beam, and the detection distance, and from the difference from the value before the turn or before the travel. An odometry (wheel distance meter) parameter adjustment method for the autonomous traveling vehicle, wherein the odometry (wheel distance meter) parameter of the autonomous traveling vehicle is adjusted.   The difference between the actual position of the autonomous traveling carriage and the goal position is calculated from the difference between the starting position and the distance to the flat plate measuring jig measured after traveling, and the odometry during the straight traveling of the autonomous traveling carriage is calculated. The parameter is calculated, The odometry (wheel distance meter) parameter adjustment method of the autonomous traveling vehicle described in claim 10.   The difference between the actual turning angle and the turning command angle of the autonomous traveling vehicle is calculated from the difference between the distance and the angle to the flat plate measuring jig measured before and after the turning, and the odometry parameter when turning the autonomous traveling vehicle The odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle according to claim 10 or 11, wherein:   Positioning treatment for setting the position of the autonomous traveling carriage at the start position or the goal position by bringing the wheels into contact with the position shifted by the wheel radius of the autonomous traveling carriage from the start position or the goal position. An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle according to any one of claims 1 to 12, wherein the parameter adjustment method is performed using a tool. In an autonomous traveling vehicle that travels with at least two traveling wheels each having an independent drive source, the travel distance and the travel direction are determined by the amount of rotation of the travel wheels and the difference in the amount of rotation of each travel wheel. An odometry function (movement distance measurement function) for estimating, and adjusting an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating the movement distance and the movement direction; An odometry (wheel distance meter) parameter adjustment method for an autonomous traveling vehicle,
A position / orientation detection sensor for detecting the position and orientation of the autonomous traveling vehicle is provided in the traveling space of the autonomous traveling vehicle, and a substantially perpendicular posture measuring jig is attached to the autonomous traveling vehicle, so that the position / orientation detection sensor is installed. The autonomous traveling vehicle is installed at a position so that the distance from each surface of the jig for posture measurement to the sensor is a predetermined distance, and the position / posture is measured. When the command is given and the running error is measured, the autonomous traveling carriage is installed at a specific distance from the position / posture detection sensor so that each surface of the posture measuring jig faces a predetermined direction. A straight command is given, and the difference between the angle given by the turn command from the position / posture before and after turning or after running of the posture measuring jig detected by the position / posture detection sensor, or the straight command Given And it obtains the difference between the specific distance, the autonomous traveling carriage of odometry autonomous traveling carriage, characterized by adjusting the (wheel rangefinder) parameter odometry (wheel rangefinder) parameter adjusting method.   From the travel distance of the autonomous traveling vehicle measured after traveling the specific distance and the inclination of the autonomous traveling vehicle with respect to the straight line connecting the start position and the goal position, the diameter ratio of each wheel of the autonomous traveling vehicle is calculated and calculated. The odometry (wheel distance meter) parameter of the autonomous traveling vehicle according to any one of claims 1 to 13, wherein an odometry (wheel distance meter) parameter of the autonomous traveling vehicle is adjusted. A wheel diameter and a wheel interval for estimating a moving distance and a moving direction by traveling with at least two traveling wheels and an imaging device each having an independent driving source and a means for measuring the rotation amount ( An odometry (wheel distance meter) parameter adjusting device for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by
Storage means for storing, as a template, a marker installed in the vicinity of a goal position at a specific distance from the start position of the autonomous traveling vehicle so as to be imaged by the imaging device, and an image of the marker previously captured by the imaging device at the goal position The autonomous traveling carriage is installed at the start position in the goal direction to give a travel instruction for the specific distance, or a turn command for one rotation is given at the goal position, and after the travel or turn, the imaging device Imaging the marker and measuring the rotation measured by the output of the rotation amount measuring means, or the difference between the rotation amount of the traveling wheel in traveling and the rotation amount of each traveling wheel, the template stored in the storage device, and the turning Alternatively, the travel distance and travel direction of the autonomous traveling vehicle are estimated from the comparison result with the marker image captured after traveling. Of the odometry (wheel rangefinder) odometry autonomous traveling vehicle, characterized in that a control means for adjusting the parameters (wheel rangefinder) parameter adjustment device.   The marker is composed of at least two markers, and the distance between the markers is determined by using the image of the marker imaged by the imaging device at the goal position and the distance to the marker of the autonomous traveling vehicle. The odometry (wheel distance meter) parameter adjusting device for an autonomous traveling vehicle according to claim 15, wherein the distance between the subsequent autonomous traveling vehicle and the marker can be calculated.   The odometry (wheel distance meter) parameter adjustment device for an autonomous traveling vehicle according to claim 15 or 16, wherein the marker is installed at a charging station of the autonomous traveling vehicle. To travel with at least two traveling wheels each having an independent drive source and a means for measuring the amount of rotation and sensors for detecting front and side obstacles, and to estimate the travel distance and travel direction An odometry (wheel distance meter) parameter adjusting device for an autonomous traveling vehicle that adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) of the vehicle,
The distance from the autonomous traveling cart installed toward the goal position that is a specific distance away from the start position to each flat plate in the measurement jig is a predetermined distance. The measurement jig is installed so that the distance to each flat plate measured in advance by the front and side obstacle detection sensors is stored, and when the turning angle error is measured, one turn is turned as it is. When calculating a running error, the measurement jig is installed at the goal position to give a running command for the specific distance, and the measurement after turning or running by the front and side obstacle detection sensors. A control hand that measures the distance to each flat plate of the jig and adjusts the odometry (wheel distance meter) parameter of the autonomous traveling vehicle from the difference from the value measured before turning or traveling stored in the storage means DOO odometry autonomous traveling vehicle characterized by comprising a (wheel rangefinder) parameter adjustment device. For obstacle detection comprising at least two beams having an angular difference capable of measuring the traveling direction inclination of at least two traveling wheels and autonomous traveling carts each having an independent drive source and a means for measuring the rotation amount Odometry (wheel) of an autonomous traveling vehicle that travels with a sensor and adjusts an odometry (wheel distance meter) parameter determined by a wheel diameter and a wheel interval (tread) for estimating a moving distance and a moving direction Distance meter) parameter adjusting device,
A flat plate measuring jig capable of reflecting the beam in the obstacle detection sensor, and a distance from the autonomous traveling carriage installed toward a goal position a specific distance away from the start position to the flat plate measuring jig is predetermined. The flat plate measuring jig is installed so that the distance is the same, and the detection distance calculated autonomously from the distance to the flat plate measuring jig and the irradiation angle of each beam by each beam in the obstacle detection sensor in advance A storage means for storing a traveling carriage attitude angle; a turn command for one rotation as it is at the time of turning angle error measurement; and a flat plate measuring jig installed at the goal position to calculate a running error; The detection distance calculated from the distance to the flat plate measuring jig and the irradiation angle of each beam after turning or traveling by the obstacle detection sensor and the autonomous traveling platform And a control means for adjusting an odometry (wheel distance meter) parameter of the autonomous traveling vehicle from a difference between a posture angle and a value measured before turning or traveling stored in the storage means. Odometry (wheel distance meter) parameter adjustment device for traveling cart.   Positioning in which the wheels are brought into contact with a position shifted from the start position or goal position by the wheel radius of the autonomous traveling carriage for installation at the start position or goal position of the autonomous traveling carriage The odometry (wheel distance meter) parameter adjusting method for an autonomous traveling vehicle according to any one of claims 15 to 19, characterized by comprising a remedy. The vehicle has at least two traveling wheels each having an independent drive source and a means for measuring the rotation amount, and has a wheel diameter and a wheel interval (tread) for estimating a moving distance and a moving direction. An odometry (wheel distance meter) parameter adjustment device for an autonomous traveling vehicle that adjusts a determined odometry (wheel distance meter) parameter,
A position / posture detection sensor for detecting the position and posture of the autonomous traveling vehicle provided in the traveling space of the autonomous traveling vehicle, a substantially right-angle posture measuring jig attached to the autonomous traveling vehicle, and the position / posture detection Storage means for storing the result of measuring the position / orientation of the autonomous traveling carriage installed at a sensor position so that the distance from each surface of the attitude measuring jig to the sensor is a predetermined distance, and measuring a turning error When the vehicle travels, a turn command for one rotation is given as it is, and when the travel error is measured, the autonomous traveling vehicle is such that each surface of the posture measuring jig faces a predetermined direction at a specific distance from the position / posture detection sensor. Is installed to give a straight-ahead command for the specific distance, and the position of the autonomous traveling vehicle is determined based on the position / posture before and after turning of the posture measuring jig detected by the position / posture detection sensor. Distance between the odometry (wheel rangefinder) for estimating a moving direction odometry autonomous traveling vehicle, characterized in that a control means for adjusting the parameters (wheel rangefinder) parameter adjustment device.

Images