Vehicle control device

Abstract

To autonomously determine operation according to a situation of a worker so as to ensure the safety of the worker and prevent work efficiency from deteriorating when the worker enters a travel route of a traveling vehicle.SOLUTION: A vehicle controller 20 for controlling an autonomously traveling vehicle 1, includes: a person detection unit 44 for detecting a person 52 on a route of the vehicle 1; an evacuation time prediction unit 45 for predicting an evacuation time required for the person 52 to evacuate from the route; an operation plan generation unit 40 for generating an operation plan including determination of the route to a destination of the vehicle 1; and a vehicle control unit 41 for controlling the vehicle according to the generated operation plan. The operation plan generation unit 40 generates a detour route for traveling to the destination by detouring the route, calculates a time required to reach the destination when traveling along the detour route, and determines whether to wait until the person 52 leaves the route or change the route to the detour route based on a calculated result and the predicted evacuation time predicted by the evacuation time prediction unit 45.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a vehicle control device that controls the operation of an autonomously traveling vehicle such as an automatic guided vehicle, an autonomous forklift, or an autonomously traveling guided robot. [Background technology]

[0002] Patent Document 1 is known as background technology in this technical field. Patent Document 1 describes an issue of "determining an autonomous operation that takes into consideration the safety of workers and surrounding conditions and does not reduce work efficiency during autonomous driving work," and describes a solution to this issue as follows: "An autonomous driving work device capable of performing autonomous driving and cleaning, which autonomously drives and cleans based on a pre-stored cleaning plan, inputs one or more pieces of surrounding information during the execution of autonomous driving and cleaning using a surrounding information input unit, and generates and executes an alternative operation to drive and / or clean based on the cleaning plan according to an input pattern consisting of the input timing and / or combination of one or more pieces of surrounding information using an alternative operation generation unit." [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2020-181434 A Summary of the Invention [Problem to be solved by the invention]

[0004] In the method described in Patent Document 1, a traveling vehicle that operates autonomously in a work area where workers and others coexist acquires surrounding information such as human gestures, speech, and work instruction signs, and autonomously generates and executes alternative actions according to the surrounding information. When a worker on the vehicle's travel route gives instructions such as "wait a moment" or "make a detour" by voice or gesture, the vehicle generates and executes an action according to the given instruction according to a motion pattern that has been memorized or learned in advance.

[0005] When multiple vehicles are operated in the same work area, the vehicles may interfere with each other due to overlapping driving paths. If a worker who does not know the positions and driving paths of multiple vehicles gives a detour instruction to a vehicle and the vehicle follows the instruction, the vehicle may interfere with the driving path of other vehicles at the detour destination, requiring a long stop, which may reduce work efficiency. In addition, if a worker gives a "wait a moment" instruction to a vehicle and the vehicle waits until the worker gives permission to resume driving, the time required for the vehicle to complete the work increases by the waiting time. In this case, if a detour path with less efficiency reduction is available, it may be possible to complete the work in a shorter time by not following the worker's instruction and taking the detour.

[0006] From the above, it can be said that instructions from workers to vehicles to detour / wait, etc., can actually be a factor that reduces the work efficiency of vehicles. In order to prevent such a decrease in efficiency, workers need to fully understand the structure of the work area, the positions of multiple vehicles, driving routes, etc., and give appropriate instructions to the vehicles, but this places a heavy burden on the workers.

[0007] In response to the above-mentioned problems, the present invention aims to enable a traveling vehicle that performs work autonomously in an environment in which workers coexist to autonomously determine its action according to the worker's situation when the worker enters the traveling path of the traveling vehicle, so as to ensure the safety of the worker while not reducing work efficiency. [Means for solving the problem]

[0008] The problem can be solved by a vehicle control device that controls a vehicle that travels autonomously along a path, the vehicle control device comprising: a person detection unit that detects people on the vehicle's path; an evacuation time prediction unit that predicts the evacuation time required for the person to evacuate from the path; an operation plan generation unit that generates an operation plan for the vehicle including route determination to a destination; and a vehicle control unit that controls the vehicle in accordance with the generated operation plan, the operation plan generation unit generating a detour route that detours around the path to travel to the destination, and calculating the time required to reach the destination if traveling along the detour route, and determining whether to wait until the person has evacuated from the path or to change the path to the detour route based on the calculation result and the evacuation time predicted by the evacuation time prediction unit. Effect of the Invention

[0009] According to the present invention, in a moving vehicle that operates autonomously in an environment where workers coexist, the time required for a person in the path of the moving vehicle to evacuate is obtained, and the vehicle autonomously decides whether to wait, detour, or resume driving, making it possible to prevent a decrease in efficiency that occurs when avoiding workers. Further features related to the present invention will become apparent from the description of the present specification and the accompanying drawings. Furthermore, the objects, configurations and effects other than those described above will become apparent from the following description of the embodiments. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a diagram showing an example of the configuration of an autonomous driving vehicle in first and second embodiments of the present invention. [Diagram 2] FIG. 2 is a diagram showing an example of the configuration of a control system for an autonomous vehicle in the first and second embodiments of the present invention. [Diagram 3] FIG. 2 is a block diagram showing the processing of the vehicle control device according to the first embodiment of the present invention. [Figure 4] FIG. 2 is a diagram showing a driving route of an autonomous vehicle in the first embodiment of the present invention. [Diagram 5] FIG. 4 is a diagram showing a process flow executed by the vehicle control device after a human detection unit detects a human on the route according to the first embodiment of the present invention. [Figure 6]FIG. 11 is a block diagram showing the processing of a vehicle control device according to a second embodiment of the present invention. [Figure 7] FIG. 11 is a diagram showing a driving route of an autonomous vehicle in a second embodiment of the present invention. [Figure 8] FIG. 11 is a diagram showing a process flow executed by the vehicle control device after the human detection unit detects a human on the route in the second embodiment of the present invention. [Figure 9] FIG. 11 is a block diagram showing the processing of a vehicle control device according to a third embodiment of the present invention. [Figure 10] FIG. 11 is a diagram showing a driving route of an autonomous vehicle in a third embodiment of the present invention. [Figure 11] FIG. 11 is a diagram showing a process flow executed by the vehicle control device after the human detection unit detects a human on the route in the third embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] [Example 1] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The following embodiment is a specific example suitable for the present invention, and the technical scope to which the present invention can be applied is not limited to these aspects. For example, the following embodiment describes an embodiment of the present invention using an autonomously traveling forklift as an example, but the present invention can be applied to transportation equipment such as AGVs (Automated Guided Vehicles) and AMRs (Autonomous Mobile Robots), as well as general equipment that travels autonomously in an environment where people coexist.

[0012] FIG. 1 shows an example of the configuration of an autonomous vehicle 1 (hereinafter, sometimes referred to as vehicle 1) according to an embodiment of the present invention. The autonomous vehicle 1 has a traveling mechanism such as tires 2, and a working mechanism consisting of forks 3, masts 4, etc. for performing a predetermined task. A front sensor 5 and a rear sensor 6 installed on the vehicle 1 detect workers and obstacles around the vehicle 1. By installing a panel 7, indicator lights 8, and a speaker 9, it is possible to notify the surroundings of the state of the vehicle 1 (driving, loading and unloading operation in progress, etc.) and the operation schedule (start of driving, backing up, turning, etc.). In addition, by installing a microphone 10 and a camera 11, it is possible to observe the voices and behaviors of workers around the vehicle 1 and the state around the vehicle 1. In addition, it is possible to directly input instructions to the vehicle 1 by operating a touch panel 12.

[0013] FIG. 2 shows an example of the configuration of a control system for the autonomous vehicle 1. The vehicle control device 20 is composed of a CPU 21, a memory 22, an input / output processing unit 23, a communication unit 24, and the like, and generates commands for driving the autonomous vehicle 1 and executing operations. A traveling mechanism PLC (Programmable Logic Controller) 26 that controls the tires 2 and brakes 25 and an operating mechanism PLC 27 that controls the forks 3 and mast 4 execute control based on commands from the vehicle control device 20. Data obtained from the aforementioned surrounding detection means 28 (front sensor 5, rear sensor 6) and information input means 29 (microphone 10, camera 11, touch panel 12) are transmitted to the vehicle control device 20. Data output by the aforementioned information output means 30 (panel 7, indicator light 8, speaker 9) is transmitted from the vehicle control device 20. The vehicle control device 20 communicates with a control server 32 via a communication module 31 to receive operation instructions. The control server 32 can also obtain the positions and driving routes of other vehicles 33 other than the vehicle 1, the status of work execution, and the position information and status of workers obtained from infrastructure sensors 34 installed in the work area and beacons 35 carried by workers.

[0014] An overview of the processing performed by the vehicle control device in the first embodiment of the present invention is shown in Fig. 3, and a travel route of the autonomous vehicle 1 is shown in Fig. 4. The motion plan generating unit 40 determines the travel route of the vehicle 1 and generates a motion plan in accordance with a work instruction given from the control server 32. For example, as shown in Fig. 4, when an instruction to pick up luggage at a destination 50 is given as a work command from the control server 32, the motion plan generating unit 40 generates a travel route 51 to the destination 50. The vehicle control unit 41 controls the traveling mechanism 42 and the working mechanism 43 in accordance with the motion plan generated by the motion plan generating unit 40.

[0015] The person detection unit 44 detects a person (a worker working in a work area) 52 present on the path of the vehicle 1 using observation data obtained from the surrounding detection means 28. The vehicle control unit 41 detects the person 52 and stops the vehicle to prevent a collision. The evacuation time prediction unit 45 predicts the evacuation time required for the person 52 to evacuate from the path of the vehicle. Using the evacuation time predicted by the evacuation time prediction unit 45, the operation plan generation unit 40 compares the time required to reach the destination 50 in the case of waiting until the person 52 has evacuated and continuing to travel on the current travel route 51 with the time required to reach the destination 50 in the case of changing the travel route to a detour route 53 and traveling, and determines whether to wait or take a detour so as to shorten the required time.

[0016] The vehicle control unit 41 controls the vehicle according to the motion plan corrected by the motion plan generating unit 40. The time required for the person 52 on the path to evacuate, i.e., the time the vehicle 1 must stop and wait, is predicted, and if the waiting time is expected to be long, the vehicle 1 takes a detour, thereby reducing the stopping time of the vehicle 1 and improving work efficiency. Here, if the person 52 evacuates immediately after the vehicle 1 approaches, or if the person 52 evacuates before the predicted evacuation time has elapsed, the vehicle 1 may immediately resume traveling. This makes it possible to reduce unnecessary waiting time.

[0017] The evacuation time prediction unit 45 uses information such as the position, walking speed, and work content of the person 52, as well as information such as the position, size, weight, characteristics, number, and status of objects around the person 52. Examples of information about a person include when the person is near a shelf where luggage is stored (position), when the person is standing still (walking speed), or when the person is carrying luggage or pushing a cart (work content), and it can be predicted that the person is working and will require a certain amount of time to evacuate.

[0018] On the other hand, when a person is in a wide passage with no objects around (position), or when the person is walking (walking speed), it can be predicted that the person is moving or not doing any particular work and can evacuate immediately. In addition, as examples of information about objects around a person, when there is luggage away from the person (position), there is large luggage around the person (size), there is heavy goods around the person (weight), there is luggage around the person that requires careful handling such as medicines or dangerous materials (characteristics), there are many pieces of luggage scattered around the person (number), there is unpacked luggage around the person (state), etc., it can be predicted that it will take time to remove those luggage and that the person cannot evacuate immediately. On the other hand, when there is no such luggage, it can be predicted that the person can evacuate in a short time. Such information about objects around a person can be stored in association with the work the person is doing.

[0019] In the evacuation request presentation unit 46, upon detection of a person 52 by the person detection unit 44, information (evacuation request) 54 is presented to the person 52 using the information output means 30, requesting the person 52 to evacuate from the route. When the speaker 9 is used, a voice such as "Please evacuate" is output. Presenting a request by voice is effective for people who do not see the vehicle. When the panel 7 is used, a message such as "Please evacuate" is displayed on the screen. When the indicator light 8 is used, a pattern corresponding to an evacuation request is displayed from among the display patterns of the indicator light that are known in advance. Evacuation requests using message display or indicator lights are effective in noisy places or for people wearing earplugs.

[0020] It is also possible to use a projector to display a message such as "Please evacuate" on the road surface or to illustrate the recommended direction for the person to evacuate with an arrow or the like. If the worker carries a beacon 35 to measure the worker's position information, the beacon 35 may be equipped with a voice output function, a screen display function, or the like to notify the request. Notification of a request by the beacon 35 is effective for a person who is far away and therefore difficult to request by voice or display from the vehicle. By actively requesting the person 52 on the path of the vehicle 1 to evacuate using the above means, it is possible to prevent the person 52 from noticing the approach of the vehicle and delaying the evacuation, thereby improving the work efficiency of the vehicle 1.

[0021] The notification unit 47 notifies 55 the person 52 on the path of the result of the decision made by the operation plan generation unit 40 to wait / detouring / restart driving, using the information output means 30. As with the request presentation by the evacuation request presentation unit 46 described above, the notification method can be a voice from the speaker 9, a display from the panel 7, the indicator light 8, a projector, or a beacon 35. If the vehicle 1 waits until the person 52 evacuates, a message such as "Will wait until evacuation" or a corresponding display pattern is output. If the vehicle 1 detouring, a message such as "Will detouring" or a corresponding display pattern is output. If driving is resumed, a message such as "Start driving" or a corresponding display pattern is output. By notifying the person 52 on the path of the future operation of the vehicle 1 in advance, the person 52 can avoid a collision or other risk that occurs when the person 52 cannot predict the operation of the vehicle 1.

[0022] If the action plan generating unit 40 determines to wait, and the person 52 does not evacuate even after the evacuation time predicted by the evacuation time predicting unit 45 has elapsed, the evacuation request presenting unit 46 issues the evacuation request 54 again. If the person 52 is late in evacuating because he or she does not notice the presentation of the evacuation request 54 or the notification 55 to wait, this directly leads to a decrease in the work efficiency of the vehicle 1. Therefore, by reissuing the request and urging the person 52 to evacuate if the person 52 is late in evacuating, a decrease in work efficiency is prevented.

[0023] 5 is a diagram showing a process flow executed by the vehicle control device 20 after the person detection unit 44 detects a person 52 on the path. In step 60, the vehicle control unit 41 controls the running mechanism 42 to stop the running of the vehicle 1. In step 61, the evacuation request presentation unit 46 outputs an evacuation request 54.

[0024] In step 62, the evacuation time prediction unit 45 predicts the time required for the person 52 to evacuate. Using a prediction model 63 created and recorded in advance by simulation, machine learning, or the like, the evacuation time of the person 52 is predicted using information such as the position and movement of the person 52 and information 64 on objects present around the person as input.

[0025] In step 65, the motion plan generating unit 40 uses the predicted evacuation time of the person 52 to calculate the time required to reach the destination 50 if the vehicle waits until the person 52 has evacuated and then resumes traveling.

[0026] In step 66, a route that takes the shortest time to reach the destination 50 among the detour routes that avoid the location where the person 52 is present is generated, and the required time is calculated. At this time, map information of the work area, the positions of the other vehicles 33, the travel routes, the work execution status, etc. 67 are obtained from the control server 32, and a detour route 53 that does not interfere with the other vehicles 33 is generated. As shown in FIG. 4, since the other vehicles 33 are scheduled to travel at the same time that the vehicle 1 travels on the detour routes 53(a) and (b), a detour route 53(c) on which the other vehicles 33 do not travel is generated. When determining whether to detour / wait in a step described later, by comparing the required time to reach the destination 50 using the detour route 53(c) that does not interfere with the travel of the other vehicles 33, it becomes possible to make a decision while taking into consideration the decrease in efficiency caused by the influence of the other vehicles 33.

[0027] In step 68, the time required to wait for the person 52 to evacuate is compared with the time required to travel along the detour route 53(c). If it is determined in step 69 that the time required to travel by detouring is shorter, in step 70, the operation plan is amended to travel along the detour route 53(c), and a detouring notification is output to the person 52.

[0028] In step 72, vehicle control unit 41 resumes driving using the detour route. If waiting will result in a shorter time, in step 73, the vehicle waits until person 52 evacuates, and outputs a wait notification 74 to person 52. If in step 75 it is determined that person 52 has evacuated, driving is resumed. If in step 76 it is determined that the evacuation time predicted in step 62 has elapsed without person 52 evacuating, in step 77 evacuation request presentation unit 46 again outputs evacuation request 54.

[0029] According to the above-described first embodiment, the time required for a person on the path to evacuate is predicted, and a detouring / waiting and resuming driving are determined so as to shorten the time required to reach the destination, thereby avoiding or shortening the waiting time until the person is evacuated, and reducing the decrease in work efficiency caused by a person entering the vehicle's driving path. At this time, the positions and driving paths of other vehicles are also considered, thereby preventing the decrease in efficiency caused by interference with other vehicles at the detouring destination. Furthermore, by actively encouraging a person on the path to evacuate using voice and display, delays in the person's evacuation action are prevented. In addition, by notifying a person of the vehicle's action when the vehicle resumes driving, it becomes possible for the person to predict the vehicle's action and prevent accidents such as contact.

[0030] [Example 2] An overview of the processing performed by the vehicle control device 20 in the second embodiment of the present invention is shown in Fig. 6, and a driving route of the autonomous vehicle 1 is shown in Fig. 7. The processing performed by the person detection unit 44, the vehicle control unit 41, and the notification unit 47 is the same as that performed in the first embodiment, and therefore the description thereof will be omitted.

[0031] The present embodiment differs from the first embodiment in that it further includes a save time request unit 80 and a save time acquisition unit 81.

[0032] The evacuation time request unit 80 uses the information output means 30 to output a request 82 for information regarding the time required for the person 52 to evacuate to the person 52 detected by the person detection unit 44. The request method is similar to the evacuation request presentation unit 46 and the notification unit 47 described above, and outputs a message such as "Please tell me how long it will take to evacuate" or a corresponding display pattern using the speaker 9, panel 7, beacon 35, etc.

[0033] The evacuation time acquisition unit 81 acquires from the person 52 response information 83 regarding the evacuation time that the person 52 inputs to the information input means 29 in response to the provision request 82. The response information regarding the evacuation time acquired from the person 52 does not necessarily have to be a numerical value such as "x seconds" or "x minutes", and may be in a larger granularity such as "1 to 2 minutes", in a format in which the user can select from preset options such as "large", "medium", or "small", or may be information or instructions such as "Evacuate immediately" or "Take a detour as evacuation is not possible".

[0034] When the microphone 10 is used as the information input means 29, speech from the person such as "It will take 1 to 2 minutes" or "Evacuation time: medium" is acquired as audio, and the presented evacuation time information and instructions are identified. When the touch panel 12 is used, options are displayed on the screen and the person presses them, or the person inputs numbers, etc. When the camera 11 is used, gestures, etc. made by the person 52 are acquired, and the previously associated evacuation time information and instructions are identified. When the beacon 35 is used, a button may be provided on the beacon 35 and the person may press it.

[0035] The motion plan generating unit 40 determines whether to wait, take a detour, or resume driving so as to shorten the time required to reach the destination 50, according to the information and instructions related to the evacuation time acquired by the evacuation time acquiring unit 81. If the accuracy of the evacuation time predicting unit 45 in the first embodiment is poor, the evacuation time of the person 52 on the path cannot be appropriately predicted, resulting in a decrease in work efficiency. Therefore, by acquiring information related to the evacuation time directly from the person 52 on the path as in this embodiment, it is possible to improve the efficiency of the decisions to wait, take a detour, or resume driving with a higher degree of accuracy.

[0036] In the evacuation time acquisition unit 81, if the provision of response information 83 regarding the evacuation time from the person 52 is not received for a certain period of time, the provision request 82 is executed again by the evacuation time request unit 80. As with the evacuation request 54 by the evacuation request presentation unit 46, if the provision of the response information 83 is delayed because the person 52 does not notice, this directly leads to a decrease in the work efficiency of the vehicle 1. If the provision of the response information 83 from the person 52 is delayed, the provision request 82 is executed again to urge the response information 83 regarding the evacuation time, thereby preventing a decrease in work efficiency.

[0037] 8 shows a process flow that the vehicle control device 20 executes after the person detection unit 44 detects the person 52. Steps 60, 65, 66, 68, 69, 70, 72, 73, 75, 76, and 77 are the same as those in the first embodiment. Further in this embodiment, in step 84, an evacuation time request unit 80 outputs a request 82 for providing information regarding the time required for evacuation to the person 52.

[0038] In step 86, the evacuation time acquisition unit 81 acquires response information 83 regarding the evacuation time from the person 52 as a response to step 84 ("Yes" in step 85). If response information 83 regarding the evacuation time is not obtained for a certain period of time ("No" in step 85), the process returns to step 84 and information regarding the evacuation time is requested again.

[0039] In the above-described second embodiment, information on the time required for evacuation is directly obtained from people on the path, thereby avoiding the evacuation time prediction required in the first embodiment and more reliably reducing efficiency decline. In addition, by accumulating the evacuation time obtained using this embodiment and information on people, etc., and performing machine learning, etc., it is also possible to improve the accuracy of the prediction model used in the first embodiment.

[0040] [Example 3] An overview of the process performed by the vehicle control device 20 in the third embodiment of the present invention is shown in Fig. 9, and a driving route of the autonomous vehicle 1 is shown in Fig. 10. In this embodiment, the processes performed by the person detection unit 44, the evacuation request presentation unit 46, the vehicle control unit 41, and the notification unit 47 are similar to those in the first and second embodiments, and therefore will not be described.

[0041] This embodiment differs from the first and second embodiments in that the evacuation time request unit 80 requests the control server 32, not the person 52, to provide information on the evacuation time.

[0042] In this embodiment, the evacuation time request unit 80 requests an external device such as the control server 32 to provide information regarding the time required for the person 52 to evacuate, regarding the person 52 on the path detected by the person detection unit 44. The control server 32 predicts the time required for the person 52 to evacuate, similar to the evacuation time prediction unit 45, from information regarding the position and movement of the person 52 and objects around the person obtained from the infrastructure sensor 34 and the beacon 35.

[0043] The evacuation time acquisition unit 81 acquires information on the evacuation time of the person 52 from the control server 32 as response information to the evacuation time request unit 80. When a calculation amount is required to predict the evacuation time of the person 52, there is a possibility that the resources for making the prediction on the vehicle control device 20 mounted on the vehicle 1 are insufficient. Therefore, by making the prediction on an external device such as the control server 32 and providing the result to the vehicle control device 20, it becomes possible to make the prediction without putting a strain on the calculation amount available to the vehicle control device 20.

[0044] The operation plan generating unit 40 determines whether to wait, take a detour, or resume traveling so as to shorten the time required to reach the destination 50, according to the information regarding the evacuation time acquired by the evacuation time acquiring unit 81. This is the same as in the second embodiment.

[0045] Here, consider a case where a large efficiency drop occurs in both detouring and waiting when the motion plan generated by the motion plan generating unit 40 is executed. For example, there are a case where the predicted evacuation time of the person 52 becomes a very long waiting time, a case where other vehicles 33 are present on many of the selectable detouring routes 53, and only the detouring route 53(c) where the efficiency is greatly reduced can be selected, and the like. In order to prevent delays in the completion of the ongoing work due to such a large efficiency drop, the work is transferred to another vehicle 86 that can perform the work instead of the vehicle 1. The transfer of the work can be executed via the control server 32. This makes it possible to effectively utilize multiple vehicles and prevent a large drop in work efficiency.

[0046] 11 shows a process flow that the vehicle control device 20 executes after the human detection unit 44 detects the human 52. Steps 60, 65, 66, 68, 69, 70, 72, 73, 74, 75, 76, and 77 are the same as those in the first and second embodiments.

[0047] Further in this embodiment, in step 90, the evacuation time request unit 80 outputs a request 91 for information on the time required for the person 52 to evacuate to the control server 32. In step 92, the evacuation time acquisition unit 81 acquires information 93 on the evacuation time from the control server 32 as a response to step 92.

[0048] If the required time to destination 50 calculated in steps 65 and 66 exceeds a certain time for both detour / waiting ("No" in step 94), the motion plan generating unit 40 determines whether there is another vehicle 86 that can perform the work in place of the host vehicle 1, and if there is another vehicle ("Yes" in step 95), in step 96 the motion plan generating unit 40 transfers the work being performed by the host vehicle 1 to the other vehicle 86. Then, in step 97, the motion plan generating unit 40 generates a motion plan for withdrawing from the current location or for performing the next scheduled work, and resumes traveling of the vehicle in step 72.

[0049] In the above-described third embodiment, by enabling prediction of evacuation time by an external device such as a server, instead of on the vehicle, it is expected that the processing load of the vehicle control device can be reduced and prediction accuracy can be improved. Furthermore, when the execution of the operation plan generated by the operation plan generating unit 40 causes a very large decrease in efficiency, it is possible to suppress the decrease in efficiency of the entire system by searching for another vehicle to which the execution can be transferred and transferring the operation.

[0050] According to the embodiment of the present invention described above, the following advantageous effects are obtained. (1) A vehicle control device according to one embodiment of the present invention includes a person detection unit that detects people on the vehicle's path, an evacuation time prediction unit that predicts the time required for the person to evacuate from the path, an operation plan generation unit that generates an operation plan for the vehicle including determining a route to a destination, and a vehicle control unit that controls the vehicle in accordance with the generated operation plan, wherein the operation plan generation unit generates a detour route to travel to the destination by detouring the path, and calculates the time required to reach the destination if traveling along the detour route, and determines whether to wait until the person has removed themselves from the path or to change the path to the detour route based on the calculation result and the evacuation time predicted by the evacuation time prediction unit.

[0051] With the above configuration, in a moving vehicle that works autonomously in an environment where workers coexist, the vehicle can obtain the time required for a person in the path of the moving vehicle to evacuate and autonomously decide whether to wait, detour, or resume driving, thereby preventing a decrease in efficiency that occurs when avoiding workers.

[0052] (2) If the person evacuates from the path earlier than the predicted evacuation time, the motion plan generation unit changes the motion plan and determines to continue traveling on the path. This makes it possible to respond quickly and appropriately even if the predicted time is not calculated accurately, such as when the prediction model used by the evacuation time prediction unit is low in accuracy or when the person's work is completed earlier than expected.

[0053] (3) The evacuation time prediction unit calculates the predicted evacuation time based on information about the person, including the person's position, walking speed, or work content. By storing this information in advance, various values ​​can be used as parameters, improving the prediction accuracy.

[0054] (4) An evacuation request presentation unit is further provided that presents an evacuation request to a person. With this, if a person remains on the route after the predicted evacuation time has elapsed, the person can be made aware of the passage of the predicted elapsed time and the presence of a vehicle, and can be encouraged to evacuate quickly.

[0055] (5) The motion plan generating unit uses map information of the work area in which the vehicle can travel and / or information on other vehicles, including the positions of other vehicles traveling in the work area, planned travel routes, work execution status, and work execution schedule, to determine whether to wait until the person has left the path or to change the path to a detour. As a result, by storing this information in advance, similar to (3), various values ​​can be used as parameters, improving prediction accuracy.

[0056] (6) A notification unit is further provided that notifies a person of the motion plan generated by the motion plan generation unit. This enables a person working to understand how the vehicle will behave in the future, and provides various options, such as considering changing or adding to the work if the vehicle will detour, or omitting part of the work and quickly finishing the work if the vehicle will wait.

[0057] (7) In addition, a vehicle control device according to another embodiment of the present invention includes a person detection unit that detects a person on the vehicle's path, an evacuation time request unit that requests information regarding the evacuation time required for the person to evacuate from the path from the person or an external device that is communicatively connected to the vehicle control device and can predict the evacuation time required for the person to evacuate from the path, an evacuation time acquisition unit that acquires response information including information regarding the evacuation time from the person or the external device to the evacuation time request unit, an operation plan generation unit that generates an operation plan for the vehicle including route determination to a destination, and a vehicle control unit that controls the vehicle in accordance with the generated operation plan, wherein the operation plan generation unit generates a detour route for traveling to the destination by detouring the path, calculates the time required to reach the destination if traveling along the detour route, and determines whether to wait until the person has evacuated from the path or to change the path to the detour route based on the result of the calculation and the response information.

[0058] With the above configuration, by directly acquiring information on the time required for evacuation from people on the path, the evacuation time prediction required in the first embodiment is avoided, and the efficiency drop is more reliably reduced. In addition, by accumulating the evacuation time obtained using this embodiment and information on people, etc., and performing machine learning, etc., it is also possible to improve the accuracy of the prediction model used in the first embodiment.

[0059] (8) If the person does not evacuate or the person does not provide information on the evacuation time within a predetermined time, the evacuation time request unit reissues the request. This prevents time from being wasted if, for example, the person does not notice the initial request.

[0060] (9) When the evacuation time acquired by the evacuation time request unit from a person or an external device and the time required to reach the destination when traveling along a detour route exceed a predetermined threshold, the motion plan generation unit determines whether there is another vehicle to which the work being performed by the vehicle can be transferred within the work area in which the vehicle is traveling. In this way, when the execution of the generated motion plan would cause a very large decrease in efficiency, it is possible to suppress the decrease in efficiency of the entire system by searching for another vehicle to which the execution can be transferred and transferring the operation.

[0061] The present invention is not limited to the above-described embodiments, and various modifications are possible. For example, the above-described embodiments have been described in detail to clearly explain the present invention, and the present invention is not necessarily limited to an embodiment having all of the configurations described. It is also possible to replace a part of the configuration of one embodiment with the configuration of another embodiment. It is also possible to add the configuration of another embodiment to the configuration of one embodiment. It is also possible to delete a part of the configuration of each embodiment, or to add or replace another configuration. [Explanation of symbols]

[0062] 1: autonomous vehicle, 20: vehicle control device, 28: surrounding detection means, 29: information input means, 30: information output means, 31: communication module, 32: control server, 33: other vehicles, 40: operation plan generation unit, 41: vehicle control unit, 44: person detection unit, 45: evacuation time prediction unit, 46: evacuation request presentation unit, 47: notification unit, 50: destination, 51: driving route, 52: person on route, 53: detour route, 54: evacuation request, 55: notification, 80: evacuation time request unit, 81: evacuation time acquisition unit, 83: response information, 93: evacuation time information, 86: other vehicles to which work can be transferred

Claims

1. A vehicle control device for controlling a vehicle that autonomously travels along a course, a person detection unit for detecting a person on a path of the vehicle; an evacuation time request unit that is communicatively connected to the vehicle control device and that requests information regarding the evacuation time required for the person to evacuate from the path from an external device that is communicatively connected to the vehicle control device and that is capable of predicting the evacuation time required for the person to evacuate from the path or from the person; an evacuation time acquisition unit that acquires, from the person or the external device, response information including information regarding the evacuation time in response to the evacuation time request unit; an operation plan generating unit that generates an operation plan for the vehicle, the operation plan including route determination to a destination; a vehicle control unit that controls the vehicle in accordance with the generated operation plan, the operation plan generation unit generates a detour route for moving to the destination by detouring the course, calculates a required time to reach the destination when moving along the detour route, and determines, based on a result of the calculation and the response information, whether to wait until the person has evacuated from the course or to change the course to the detour route; When the evacuation time acquired by the evacuation time request unit from the person or the external device and the time required to reach the destination when traveling along the detour route exceed a predetermined threshold, the operation plan generation unit determines whether there is another vehicle to which the work being performed by the vehicle can be transferred within a work area in which the vehicle is traveling. A vehicle control device comprising:

2. The vehicle control device according to claim 1, when the evacuation of the person or the provision of information on the evacuation time by the person is not performed within a predetermined time, the evacuation time request unit executes the presentation of the request again. A vehicle control device comprising:

Images