Mobile body evaluation system, mobile body evaluation method, mobile body evaluation device, and mobile body evaluation program

Abstract

Provided is a mobile body evaluation system comprising: an acquisition means for acquiring sensor information resulting from sensing the environment of a mobile body during traveling and history regarding the traveling position of the mobile body; a calculation means for calculating an assumed route assumed for the mobile body in the environment, on the basis of the sensor information and the history regarding the position; a prediction means for predicting the traveling route of the mobile body controlled on the basis of the assumed route; and an evaluation means for evaluating the traveling of the mobile body on the basis of the assumed route and the traveling route. The assumed route is a route in which the mobile body can travel while avoiding an obstacle.

Description

MOBILE BODY EVALUATION SYSTEM, MOBILE BODY EVALUATION METHOD, MOBILE BODY EVALUATION DEVICE, AND MOBILE BODY EVALUATION PROGRAM

[0001] The present disclosure relates to a mobile object evaluation system, a mobile object evaluation method, a mobile object evaluation device, and a mobile object evaluation program.

[0002] The cost of travel for a mobile object is calculated to determine an efficient route. In Patent Document 1, an optimal route combination is determined from among multiple candidate routes for a mobile object.

[0003] Japanese Patent Application Laid-Open No. 2003-44998

[0004] The technology of Patent Document 1 determines an optimal route combination. However, even if a mobile object travels along the optimal route at the time of calculation, this does not necessarily mean that the mobile object is traveling optimally. Therefore, an object of the present disclosure is to provide a mobile object evaluation system, a mobile object evaluation method, and a mobile object evaluation device that evaluate the travel of a route taken by a mobile object.

[0005] The mobile object evaluation system of the present disclosure is a mobile object evaluation system that includes: an acquisition means for acquiring sensor information that senses the environment in which a mobile object is traveling and a history of the traveling position of the mobile object; a calculation means for calculating an expected path of the mobile object in the environment based on the sensor information and the position history; a prediction means for predicting a traveling path of the mobile object controlled based on the expected path; and an evaluation means for evaluating the traveling of the mobile object based on the expected path and the traveling path.

[0006] The mobile body evaluation method disclosed herein acquires sensor information that senses the environment in which the mobile body is traveling and a history of the location of the traveling mobile body, calculates a predicted path of the mobile body in the environment based on the sensor information and the location history, predicts a traveling path of the mobile body controlled based on the predicted path, and evaluates the traveling of the mobile body based on the predicted path and the traveling path.

[0007] The mobile body evaluation device of the present disclosure is a mobile body evaluation device that includes: an acquisition means for acquiring sensor information that senses the environment while the mobile body is traveling and a history of the traveling position of the mobile body; a calculation means for calculating an expected path of the mobile body in the environment based on the sensor information and the position history; a prediction means for predicting a traveling path of the mobile body controlled based on the expected path; and an evaluation means for evaluating the traveling of the mobile body based on the expected path and the traveling path.

[0008] The present disclosure provides a mobile object evaluation system, a mobile object evaluation method, a mobile object evaluation device, and a mobile object evaluation program for evaluating the travel of a route traveled by a mobile object.

[0009] FIG. 1 is a block diagram showing a configuration of a moving body evaluation system according to the present disclosure. FIG. 2 is a flowchart of a moving body evaluation method according to the present disclosure. FIG. 3 is a block diagram showing a configuration of a moving body evaluation device according to the present disclosure. FIG. 4 is an exemplary relationship diagram of time and speed when controlling a moving body according to the present disclosure. FIG. 5 is a diagram showing an example of an assumed path and a predicted driving path of a moving body according to the present disclosure. FIG. 6 is a block diagram showing a detailed configuration of a moving body evaluation device according to the present disclosure. FIG. 7 is a detailed flowchart of a moving body evaluation method according to the present disclosure. FIG. 8 is a block diagram showing a configuration of an information processing device according to the present disclosure.

[0010] (Description of a Mobile Object Evaluation System According to an Embodiment) Hereinafter, a configuration example of a mobile object evaluation system 100 will be described with reference to FIG. 1. The mobile object evaluation system 100 is a system that calculates an expected path and a predicted driving path of a mobile object at a certain point in time, and evaluates the operation of the mobile object based on the expected path and the driving path. The mobile object is, for example, a mobile object used in a factory, such as a forklift, an AMR (Autonomous Mobile Robot), or a work vehicle. The mobile object may be either a manually operated or automatically operated one.

[0011] As shown in FIG. 1 , the moving object evaluation system 100 includes an acquisition unit 101 , a calculation unit 102 , a prediction unit 103 , and an evaluation unit 104 .

[0012] The acquisition unit 101 acquires sensor information that senses the environment while the mobile object is traveling and a history of the traveling position of the mobile object. The sensor information that senses the environment is sensor information that senses surrounding obstacles, moving objects, etc. while the mobile object is moving. The sensor is, for example, a LiDAR (Light Detection and Ranging), a depth camera, an imaging device, etc. It is preferable that the sensor be able to measure the distance to an object. The history of the traveling position of the mobile object is the route that the mobile object actually traveled.

[0013] The calculation unit 102 calculates an estimated path of the mobile body in the environment in which the mobile body actually traveled, based on the sensor information and the position history. The estimated path is a path connecting the positions where the mobile body was located after a specific time has elapsed from a certain point in the travel history. The estimated path is a path that the mobile body can travel at a certain point in time while avoiding obstacles. The travel path may also be a path that the mobile body can travel to a target position in a short distance while avoiding obstacles. The estimated path is, for example, the shortest route to a target position at a certain point in time. However, the estimated path does not have to take the shortest route as a result of detecting obstacles, slopes, etc. The estimated path may also be interpreted as an ideal path. The ideal path is the most efficient travel path, for example, the shortest distance that can be reached in the shortest time. Route generation algorithms include Theta star planner, A* planner, ReedSepp, etc.

[0014] The certain point in time is a point in time determined at a certain interval, such as 1 second or 2 seconds after the start. For example, the route from 1 second to 5 seconds after the start is calculated as the estimated route. At 2 seconds after the start, the route up to 6 seconds after the start is calculated as the estimated route. The accuracy of evaluating the moving object increases as the interval is narrowed, for example, from 1 second to 0.5 seconds.

[0015] The prediction unit 103 predicts the travel path of the mobile object controlled based on the assumed path. The travel path is predicted by performing stochastic fluctuation simulation of the assumed path. The stochastic fluctuation simulation involves performing a simulation multiple times, taking into account stochastic fluctuations in the travel of the mobile object. For example, in the process of updating the coordinates and attitude angle of the mobile object, a process of adding a random number proportional to the travel speed is added in addition to fluctuation amounts uniquely determined depending on the steering angle and travel speed. In a simulated mobile robot operation model, an example of an update model for machine information such as position coordinates and attitude is disclosed in Zhou, Yu, and Gregory S. Chirikjian, "Probabilistic models of dead-reckoning error in nonholonomic mobile robots," 2003 IEEE International Conference on Robotics and Automation (Cat. No. 03CH37422), Vol. 2, IEEE, 2003. A pure pursuit method may also be used for path tracking control. As shown in FIG. 4, examples of speed calculation include a case where the speed becomes constant after sudden acceleration, a case where the speed becomes constant after slow acceleration, a case where the speed moves at a constant speed, a case where the speed decelerates and stops, and a case where the speed decelerates and stops suddenly.

[0016] FIG. 5 shows an example in which three types of travel paths of a moving object are simulated. As shown in FIG. 5, it is assumed that the moving object moves based on an assumed path from time t=0 to time t=T. When various acceleration and deceleration are input, travel paths represented by three lines are simulated. As shown by S=0.7, it is predicted that there will be a large deviation from the assumed path, as shown by S=0.4, there will be a slight deviation from the assumed path, as shown by S=0.1, and there will be almost no deviation from the assumed path. Here, S is the predicted cost, which will be described later.

[0017] The evaluation unit 104 evaluates the travel of the mobile object based on the assumed route and the travel route. The method of calculating the predicted cost S in each simulation is as follows: q(xｉ ) is the predicted cost S, and constants α, b, and c are used. ｉ Let v be the current velocity. ｍａｘ is the maximum speed, Δy is the deviation of the predicted travel path from the assumed path, and γ is expressed as whether or not the vehicle has entered a no-entry area. The predicted cost S is the cost of a moving body passing through the predicted travel path. The current speed is the speed of a moving body traveling through the predicted travel path. A no-entry area is an area that must not be included in the path, such as a wall or an obstacle. Obstacles include people and other moving bodies. In other words, the predicted cost S = constant (current speed - maximum speed) ２ + constant (deviation of predicted driving route from assumed route) ２ + constant (current speed, whether it is a no-entry zone) ２ It can be calculated as follows.

[0018] As described in International Publication No. 2022 / 070324, the instantaneous evaluation value J is expressed as follows, where β is a constant, E is an expected value, and S is a predicted cost: The instantaneous evaluation value J is a value that evaluates the risk when a mobile object travels along each of a plurality of assumed routes.

[0019] In the case of safe driving, where the vehicle travels at an expected speed along an expected route without entering a no-entry area, the instantaneous evaluation value J is small. If the vehicle enters a no-entry area, the instantaneous evaluation value J becomes large. In this way, the instantaneous evaluation value J is a score that takes into account not only the simple probability of an event occurring, but also the severity of the risk.

[0020] The travel of the mobile object is evaluated using one or more of the maximum value, average value, median value, and percentile value of this instantaneous evaluation value J.

[0021] In this way, a mobile object evaluation system is provided that calculates the assumed path and predicted travel path of a mobile object at a certain point in time and evaluates the travel of the path taken by the mobile object. The acquisition unit 101, the calculation unit 102, the prediction unit 103, and the evaluation unit 104 may be read as an acquisition means, a calculation means, a prediction means, and an evaluation means, respectively.

[0022] Such a mobile object evaluation system evaluates the trajectory of a mobile object through simulation and evaluates the quality of the maneuver by examining whether a high-risk situation occurs. The mobile object evaluation system can be applied to autonomous control, remote control, and on-board control.

[0023] The mobile object evaluation system of the present disclosure can notify operators of dangerous driving and prevent accidents. The mobile object evaluation system of the present disclosure can also be used to compare control algorithms for autonomously moving mobile robots.

[0024] (Description of Moving Body Evaluation Method According to an Embodiment) The moving body evaluation method will be described below with reference to FIG. 2. As shown in FIG. 2, first, sensor information and a position history are acquired (step S201). The acquisition unit 101 acquires sensor information that senses the environment in which the moving body is traveling and a history of the moving body's position. Next, an expected path is calculated (step S202). The calculation unit 102 calculates an expected path of the moving body in the environment in which the moving body is traveling based on the sensor information and the position history. Next, a traveling path is predicted (step S203). The prediction unit 103 predicts the traveling path of the moving body controlled based on the expected path. Finally, the traveling of the moving body is evaluated (step S204). The evaluation unit 104 evaluates the traveling of the moving body based on the expected path and the traveling path.

[0025] The mobile object evaluation system 100 and the mobile object evaluation method are realized by, for example, an information processing device 800. As shown in FIG. 8 , the information processing device 800 includes a processor 801 that executes a program to perform processing and a memory 802 that stores the program. Since the mobile object evaluation method is performed by executing the program, the present disclosure also discloses a mobile object evaluation program. The information processing device 800 may be configured as a single device or multiple devices. Furthermore, the information processing device 800 may be a cloud server in which some or all of its functions are distributed. When the information processing device 800 is a single device, it becomes a mobile object evaluation device 300 including an acquisition unit 301, a calculation unit 302, a prediction unit 303, and an evaluation unit 304, as shown in FIG. 3 . The acquisition unit 301, the calculation unit 302, the prediction unit 303, and the evaluation unit 304 may be interpreted as an acquisition means, a calculation means, a prediction means, and an evaluation means, respectively.

[0026] The mobile object evaluation method, mobile object evaluation program, and mobile object evaluation device disclosed herein can notify operators of dangerous driving and prevent accidents. Furthermore, the mobile object evaluation method, mobile object evaluation program, and mobile object evaluation device disclosed herein can compare control algorithms of autonomously traveling mobile robots.

[0027] (Description of Moving Object Evaluation System and Moving Object Evaluation Method According to First Embodiment) Hereinafter, a configuration example of a moving object evaluation system 600 according to the first embodiment will be described using FIG. 6. Also, a moving object evaluation method according to the first embodiment will be described using the flowchart of FIG. 7. As shown in FIG. 6, the moving object evaluation system 600 includes a sensor information storage unit 601, a no-entry area calculation unit 602, a path calculation unit 603, a probability fluctuation simulation processing unit 604, an instantaneous evaluation value calculation unit 605, an instantaneous evaluation value storage unit 606, and a maneuvering score calculation unit 607.

[0028] The sensor information storage unit 601 stores sensor information acquired in the environment in which the mobile body moves. As shown in Fig. 7, the sensor information storage unit 601 is used to reproduce the sensor information history of the mobile robot (step S701).

[0029] The no-entry area calculation unit 602 sets no-entry areas. For example, it sets areas where a moving object should not enter, such as walls, steps, obstacles, etc. If a moving object enters these areas, a large deduction will be made in the calculation of the evaluation score.

[0030] The route calculation unit 603 calculates an expected route along which the moving object will travel. As shown in Fig. 7, the route calculation unit 603 first acquires a point T1 seconds in the future as a destination point (step S702). Next, the route calculation unit 603 generates a route from the current location to the destination point (step S703). Therefore, the expected route is calculated multiple times for each time point.

[0031] The stochastic variation simulation processing unit 604 predicts a travel path multiple times, taking into account stochastic variations in the travel of the mobile object. As shown in Fig. 7, the stochastic variation simulation processing unit 604 performs path following control through M simulations, each using a different pseudo-random number (step S704). The stochastic variation simulation processing unit predicts a travel path for various accelerations and steering angles.

[0032] The instantaneous evaluation value calculation unit 605 calculates an instantaneous evaluation value J, which is a value for evaluating the risk when a mobile object travels multiple assumed routes. As shown in Fig. 7, the instantaneous evaluation value J is calculated using M predicted costs obtained from M simulations (step S705). In this way, since the travel route is predicted multiple times, the instantaneous evaluation value J is calculated each time.

[0033] The instantaneous evaluation value storage unit 606 stores multiple instantaneous evaluation values ​​J. At this time, as shown in FIG. 7, it is determined whether or not the maneuver has been completed (step S706). If the maneuver has not been completed (NO in step S706), the process returns to step S702. Therefore, the instantaneous evaluation value storage unit 606 stores multiple simulation results for each time point. If the maneuver has been completed (YES in step S706), the process proceeds to the next step 707.

[0034] The maneuvering score calculation unit 607 evaluates the traveling of the moving object using one or more of the maximum value, average value, median value, and percentile value of the multiple instantaneous evaluation values ​​J. As shown in Fig. 7 , the maneuvering score calculation unit 607 calculates the maneuvering score based on the instantaneous evaluation values ​​J calculated so far (step S707).

[0035] The mobile object evaluation system and mobile object evaluation method disclosed herein can notify operators of dangerous driving and prevent accidents. Furthermore, the mobile object evaluation system and mobile object evaluation method disclosed herein can compare the control algorithms of autonomously traveling mobile robots.

[0036] Furthermore, some or all of the processes in the above-described mobile object evaluation system 100, mobile object evaluation device 300, and mobile object evaluation system 600 can be realized as a computer program. Such a program can be stored on various types of non-transitory computer-readable media and supplied to a computer. Non-transitory computer-readable media include various types of tangible recording media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). Furthermore, the program may be supplied to a computer by various types of temporary computer-readable media. Examples of the temporary computer-readable medium include an electric signal, an optical signal, and an electromagnetic wave. The temporary computer-readable medium can provide the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

[0037] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above-described embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. Furthermore, each embodiment can be combined with other embodiments as appropriate.

[0038] Each drawing is merely an example for describing one or more embodiments. Each drawing may not relate to only one particular embodiment, but may also relate to one or more other embodiments. As will be understood by those skilled in the art, various features or steps described with reference to any one drawing can be combined with features or steps shown in one or more other drawings to create, for example, an embodiment not explicitly shown or described. Not all features or steps shown in any one drawing are necessary to describe an exemplary embodiment, and some features or steps may be omitted. The order of steps described in any drawing may be changed as appropriate.

[0039] Some or all of the above embodiments may be described as, but are not limited to, the following supplementary notes. (Supplementary Note 1) A mobile object evaluation system comprising: an acquisition means for acquiring sensor information that senses an environment during travel of a mobile object and a history of travel positions of the mobile object; a calculation means for calculating an expected path of the mobile object in the environment based on the sensor information and the position history; a prediction means for predicting a travel path of the mobile object controlled based on the expected path; and an evaluation means for evaluating the travel of the mobile object based on the expected path and the travel path. (Supplementary Note 2) The mobile object evaluation system according to Supplementary Note 1, wherein the expected path is a path along which the mobile object can travel while avoiding obstacles. (Supplementary Note 3) The mobile object evaluation system according to Supplementary Note 1, wherein the prediction means uses a stochastic fluctuation simulation that performs a plurality of simulations taking into account stochastic fluctuations in the travel of the mobile object. (Supplementary Note 4) The mobile object evaluation system according to Supplementary Note 1, wherein a no-entry zone is set for the mobile object, and the evaluation is lowered when the travel path is predicted to enter the no-entry zone. (Supplementary Note 5) The mobile object evaluation system according to Supplementary Note 1, wherein the travel path prediction is performed multiple times, an instantaneous evaluation value is obtained based on the assumed route and the multiple travel paths, and the travel of the mobile object is evaluated using an average, median, or percentile value of the multiple instantaneous evaluation values. (Supplementary Note 6) The mobile object evaluation system according to Supplementary Note 1, wherein the mobile object is a forklift, an AMR (Autonomous Mobile Robot), or a work vehicle. (Supplementary Note 7) A mobile object evaluation method comprising: acquiring sensor information sensing an environment during travel of the mobile object and a history of travel positions of the mobile object; calculating a predicted assumed route of the mobile object in the environment based on the sensor information and the position history; predicting a travel path of the mobile object controlled based on the assumed route; and evaluating the travel of the mobile object based on the assumed route and the travel path. (Supplementary Note 8) The mobile object evaluation method according to Supplementary Note 7, wherein the assumed route is a route that the mobile object can travel while avoiding obstacles.(Supplementary Note 9) The mobile object evaluation method according to Supplementary Note 7, wherein the prediction uses a stochastic fluctuation simulation that performs a plurality of simulations taking into account stochastic fluctuations in the traveling of the mobile object. (Supplementary Note 10) A mobile object evaluation device comprising: an acquisition means for acquiring sensor information that senses an environment during which the mobile object is traveling and a history of the traveling positions of the mobile object; a calculation means for calculating an anticipated path of the mobile object in the environment based on the sensor information and the position history; a prediction means for predicting a traveling path of a mobile object controlled based on the anticipated path; and an evaluation means for evaluating the traveling of the mobile object based on the anticipated path and the traveling path. (Supplementary Note 11) A mobile object evaluation program that causes an information processing device to execute the following steps: acquiring sensor information that senses an environment during which the mobile object is traveling and a history of the traveling positions of the mobile object; calculating an anticipated path of the mobile object in the environment based on the sensor information and the position history; predicting a traveling path of a mobile object controlled based on the anticipated path; and evaluating the traveling of the mobile object based on the anticipated path and the traveling path.

[0040] Some or all of the elements (e.g., configurations and functions) described in Supplementary Notes 2 to 6 that are dependent on Supplementary Note 1 (e.g., system) may also be dependent on Supplementary Note 7 (e.g., method), Supplementary Note 10 (e.g., device), and Supplementary Note 11 (e.g., program) in the same dependency relationship as Supplementary Note 2 to 6. Some or all of the elements described in any Supplementary Note may be applied to various hardware, software, recording means for recording software, systems, and methods.

[0041] This application claims priority based on Japanese Patent Application No. 2024-7997, filed January 23, 2024, the disclosure of which is incorporated herein by reference in its entirety.

[0042] The mobile object evaluation system of the present disclosure evaluates the travel of a route taken by a mobile object such as a forklift, an AMR, or a work vehicle.

[0043] 100 Mobile object evaluation system, 101 Acquisition unit, 102 Calculation unit, 103 Prediction unit, 104 Evaluation unit, 600 Mobile object evaluation system, 601 Sensor information storage unit, 602 No-entry area calculation unit, 603 Route calculation unit, 604 Probability fluctuation simulation processing unit, 605 Instantaneous evaluation value calculation unit, 606 Instantaneous evaluation value storage unit, 607 Maneuver score calculation unit

Claims

1. An acquisition means for acquiring sensor information that senses the environment during the travel of a moving body and the history of the travel position of the moving body; a calculation means for calculating an assumed route of the moving body in the environment based on the sensor information and the history of the position; a prediction means for predicting the travel route of the moving body controlled based on the assumed route; and an evaluation means for evaluating the travel of the moving body based on the assumed route and the travel route. A moving body evaluation system comprising the above.

2. The moving body evaluation system according to claim 1, wherein the assumed route is a route along which the moving body can travel while avoiding obstacles.

3. The moving body evaluation system according to claim 1, wherein the prediction means uses a probabilistic variation simulation that performs a plurality of simulations considering probabilistic variations in the travel of the moving body.

4. The moving body evaluation system according to claim 1, wherein an entry prohibited area for prohibiting entry of the moving body is set, and when it is predicted that the travel route enters the entry prohibited area, the evaluation is lowered.

5. The prediction of the travel route is performed a plurality of times, and based on the assumed route and the plurality of travel routes, an instantaneous evaluation value, which is a value for evaluating the risk when the moving body travels each of the plurality of assumed routes, is obtained, and the travel of the moving body is evaluated using the average value, median value, or percentile value of the plurality of instantaneous evaluation values. The moving body evaluation system according to claim 1.

6. The moving body according to claim 1 is a forklift, an AMR (Autonomous Mobile Robot), or a work vehicle. The moving body evaluation system described.

7. A moving body evaluation method for acquiring sensor information that senses the environment during the travel of a moving body and the history of the travel position of the moving body, calculating an assumed route of the moving body in the environment based on the sensor information and the history of the position, predicting the travel route of the moving body controlled based on the assumed route, and evaluating the travel of the moving body based on the assumed route and the travel route.

8. The moving body evaluation method according to claim 7, wherein the assumed route is a route along which the moving body can travel while avoiding obstacles.

9. The prediction according to claim 7 uses a probabilistic variation simulation that performs a plurality of simulations considering probabilistic variations in the travel of the moving body.

10. The mobile body evaluation method according to claim 7, wherein a prohibited entry area of the mobile body is set, and when it is predicted that the travel route enters the prohibited entry area, the evaluation is lowered.

11. The prediction of the travel route is performed multiple times, and based on the assumed route and the plurality of travel routes, an instantaneous evaluation value, which is a value for evaluating the risk when the mobile body travels each of the plurality of assumed routes, is obtained, and the travel of the mobile body is evaluated by an average value, a median value, or a percentile value of the plurality of instantaneous evaluation values. The mobile body evaluation method according to claim 7.

12. The mobile body is a forklift, an AMR (Autonomous Mobile Robot), or a work vehicle. The mobile body evaluation method according to claim 7.

13. An acquisition means for acquiring sensor information that senses the environment during the travel of the mobile body and a history of the travel position of the mobile body, a calculation means for calculating an assumed route of the mobile body in the environment based on the sensor information and the history of the position, a prediction means for predicting a travel route of the mobile body controlled based on the assumed route, and an evaluation means for evaluating the travel of the mobile body based on the assumed route and the travel route. A mobile body evaluation device comprising:

14. The assumed route according to claim 13 is a route along which the mobile body can travel while avoiding obstacles.

15. The prediction means uses a probabilistic variation simulation that performs a simulation considering probabilistic variations in the travel of the mobile body multiple times. The mobile body evaluation device according to claim 13.

16. The mobile body evaluation device according to claim 13, wherein a prohibited entry area of the mobile body is set, and when it is predicted that the travel route enters the prohibited entry area, the evaluation is lowered.

17. The prediction of the travel route is performed multiple times, and based on the assumed route and the plurality of travel routes, an instantaneous evaluation value, which is a value for evaluating the risk when the mobile body travels each of the plurality of assumed routes, is obtained, and the travel of the mobile body is evaluated by an average value, a median value, or a percentile value of the plurality of instantaneous evaluation values. The mobile body evaluation device according to claim 13.

18. The mobile body is a forklift, an AMR (Autonomous Mobile Robot), or a work vehicle. The mobile body evaluation device according to claim 13.

19. An information processing apparatus is caused to execute a moving body evaluation program that acquires sensor information sensing an environment during travel of a moving body and a history of positions of travel of the moving body, calculates an assumed route of the moving body in the environment based on the sensor information and the history of positions, predicts a travel route of the moving body controlled based on the assumed route, and evaluates the travel of the moving body based on the assumed route and the travel route.

20. The moving body evaluation program according to claim 19, wherein the assumed route is a route along which the moving body can travel while avoiding obstacles.

Images