Remote support device, remote support method, and remote support program

Abstract

To reduce the time spent waiting for support at the point of support when there is no or expected to be no operator available to support in the automated driving of the automated vehicle.SOLUTION: A remote support device includes a control unit that controls the operation of an automated vehicle so that there are always operators at the support point where remote support is provided to the automated vehicle, when there is no or expected to be no operator available to perform the remote support at the support point.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to a remote assistance device, a remote assistance method, and a remote assistance program. [Background technology]

[0002] Patent document 1 discloses a remote assistance system (100) that provides remote assistance to a vehicle (10), comprising: a route planning unit (62) that identifies a route for the vehicle to reach its destination via an assistance point where remote assistance is provided to the vehicle; and a prediction unit (66) that predicts the timing at which an operator will begin remote assistance for the vehicle based on the estimated time at which the vehicle will arrive at the assistance point included in the identified route. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-160146 Summary of the Invention [Problem to be solved by the invention]

[0004] The technology described in Patent Document 1 above predicts the number of operators required based on a vehicle route plan, making it possible to provide remote support based on a long-term (hourly) personnel plan.

[0005] However, the technology described in Patent Document 1 has a problem in that if an unexpected event or the like results in more personnel being required than predicted, the time spent waiting for support at the support point becomes long.

[0006] The present disclosure aims to provide a remote assistance device, a remote assistance method, and a remote assistance program that can shorten the time spent waiting for assistance at an assistance point when there is no operator available to assist the autonomous driving of an autonomous vehicle or when it is predicted that there will be no more available operators. [Means for solving the problem]

[0007] A remote assistance device according to a first aspect of the present disclosure includes a control unit that controls the operation of an autonomous vehicle so that an operator is not unavailable at an assistance point where remote assistance is provided to the autonomous vehicle when there is no operator available to provide the remote assistance or when it is predicted that there will be no available operators at the assistance point.

[0008] A remote assistance method according to a second aspect of the present disclosure includes a computer that controls the operation of an autonomous vehicle so that an operator is not unavailable at an assistance point where remote assistance is provided to the autonomous vehicle when there is no operator available to provide the remote assistance or when it is predicted that there will be no available operators at the assistance point.

[0009] A remote assistance program according to a third aspect of the present disclosure causes a computer to execute processing that includes controlling the operation of an autonomous vehicle so that an operator is not unavailable at an assistance point where remote assistance is provided to the autonomous vehicle when there is no operator available to provide the remote assistance or when it is predicted that there will be no available operators at the assistance point. [Effects of the Invention]

[0010] According to the present disclosure, when there is no availability of an operator to assist the autonomous driving of an autonomous vehicle or when it is predicted that there will be no availability, it is possible to shorten the time spent waiting for assistance at an assistance point. [Brief explanation of the drawings]

[0011] [Figure 1] FIG. 1 is a configuration diagram of an autonomous driving system. [Figure 2] FIG. 2 is a diagram illustrating a hardware configuration of a remote support device. [Figure 3] 10 is a flowchart of a remote support process. [Figure 4] FIG. 10 is a diagram for explaining a specific example 1 of the remote support process. [Figure 5] FIG. 10 is a diagram for explaining a specific example 1 of the remote support process. [Figure 6] FIG. 10 is a diagram for explaining a specific example 2 of the remote support process. [Figure 7] FIG. 10 is a diagram for explaining a specific example 2 of the remote support process. [Figure 8] FIG. 10 is a diagram for explaining a specific example 3 of the remote support process. [Figure 9] FIG. 10 is a diagram for explaining a specific example 3 of the remote support process. [Figure 10] FIG. 10 is a diagram for explaining a specific example 4 of the remote support process. [Figure 11] FIG. 10 is a diagram for explaining a specific example 5 of the remote support process. [Figure 12] FIG. 10 is a diagram for explaining a specific example 5 of the remote support process. [Figure 13] FIG. 10 is a diagram for explaining a specific example 6 of the remote support process. DETAILED DESCRIPTION OF THE INVENTION

[0012] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0013] As shown in Fig. 1, the autonomous driving system 100 of this embodiment includes a vehicle 10, a remote assistance device 50 located at an autonomous driving center, and an operator terminal 80 operated by an operator. The vehicle 10 in this embodiment is an autonomously driven vehicle that operates automatically. The autonomous driving system 100 is a system in which an operator at a remote location remotely assists the vehicle 10 when the vehicle 10 (autonomously driven vehicle) arrives at a predetermined assistance point.

[0014] In this embodiment, remote assistance refers to an operator operating the vehicle 10 from a remote location, or an operator monitoring the vehicle 10 from a remote location. An assistance point refers to a point where it is difficult for the vehicle 10 to drive autonomously, or a point where the operator can remotely assist the vehicle. Examples of assistance points include, but are not limited to, intersections, construction sites, parking lots, expressways, motorways, service areas, parking areas, baggage collection points, toll booths, and interchanges.

[0015] The autonomous driving system 100 of this embodiment includes one or more vehicles 10. Each vehicle 10 includes a vehicle-side communication device 12, a camera 14, a positioning device 16, and a vehicle-side control unit 18.

[0016] The vehicle-side communication device 12 performs wireless communication with the remote assistance device 50 .

[0017] The camera 14 captures the surroundings of the vehicle 10. In this embodiment, the camera 14 captures the area ahead of the vehicle 10.

[0018] The positioning device 16 includes a GPS (Global Positioning System) receiver or a GNSS (Global Navigation Satellite System) receiver, and determines the current position of the vehicle 10.

[0019] The vehicle-side control unit 18 controls the vehicle-side communication device 12, the camera 14, and the positioning device 16. The vehicle-side control unit 18 is configured as a computer equipped with a CPU and memory, and functions as an operation control unit 20 by the CPU executing a program stored in the memory.

[0020] The operation control unit 20 controls the operation (acceleration, deceleration, braking, steering) of the vehicle based on remote operation by the operator. The operation control unit 20 also transmits vehicle information, including identification information (vehicle ID) and location information of the vehicle 10, to the remote assistance device 50 repeatedly at a predetermined cycle via the vehicle-side communication device 12. The vehicle information may include information for identifying the vehicle model and occupants of the vehicle 10, and information indicating the location of the destination. When the operator provides remote assistance for the vehicle 10, the operation control unit 20 also transmits video captured by the camera 14 to the remote assistance device 50 and the operator terminal 80 via the vehicle-side communication device 12.

[0021] The remote support device 50 includes a communication unit 52, a control unit 60, and a storage device 70.

[0022] The communication unit 52 performs wireless communication with the vehicle 10. The communication unit 52 also performs wired or wireless communication with the operator terminal 80.

[0023] The storage device 70 stores a map database 72, an operator information database 74, and a remote assistance program 76.

[0024] The positions of the support points are registered in the map database 72. The map database 72 also stores road data indicating the connection status of intersections and roads, congestion information, traffic regulation information, and traffic information including traffic volume for each support point. The traffic information is obtained from another server via the communication unit 52, for example, and stored.

[0025] The operator information database 74 records, for example, the number of currently available operators among the operators who provide remote support for vehicles.

[0026] The control unit 60 controls the communication unit 52 and the storage device 70. Functionally, the control unit 60 includes a route planning unit 62, a remote support unit 64, a prediction unit 66, a recording unit 68, and an allocation unit 69.

[0027] The route planning unit 62 specifies a route for each vehicle 10 to travel. More specifically, the route planning unit 62 specifies a route for each vehicle 10 to travel from the departure point to the destination while passing through support points, based on road data, traffic information, and the positions of the support points recorded in the map database 72. The departure point, destination, and support points to be passed through are set, for example, by an operator or their manager, or in response to a request from a passenger. The route planning unit 62 distributes the specified route to each vehicle 10 via the communication unit 52. Note that the route may be specified in the vehicle-side control unit 18 of each vehicle 10. Alternatively, the route may be specified in response to a request from each vehicle 10.

[0028] The remote support unit 64 provides remote support for the vehicle 10 at the support point based on operations by the operator. For example, when the remote support unit 64 determines that the vehicle 10 has arrived at the support point based on location information transmitted from the vehicle 10, or when the remote support unit 64 receives a remote support request from the vehicle 10, the remote support unit 64 receives an operation signal from the operator to the operator terminal 80, and provides remote support for the vehicle 10 at the support point based on the operation signal. When providing remote support, the remote support unit 64 transmits vehicle information and video captured by the camera 14 provided in the vehicle 10 to the operator terminal 80.

[0029] The prediction unit 66 predicts the timing at which an operator will begin remote support for the vehicle 10, based on the estimated time at which the vehicle 10 will arrive at a support point included in the route identified by the route planning unit 62, and also predicts the time for which the remote support will be provided (hereinafter referred to as "required time"). The prediction unit 66 can, for example, learn the relationship between the past required time and traffic volume for each support point, and predict the required time from the learning results and the current traffic volume. The prediction unit 66 in this embodiment has a function of predicting the number of operators required to simultaneously support multiple vehicles, based on the estimated time at which each vehicle 10 will arrive at a support point included in the route identified for each vehicle 10.

[0030] The recording unit 68 records the history of the number of operators who provided remote assistance for each time period in the operator information database 74. The recording unit 68 also records the history of the time required for each operator to remotely assist the vehicle 10 at a support point in the operator information database 74 for each support point. The recording unit 68 also records the history of traffic volume for each support point in the operator information database 74. The recording unit 68 also calculates the probability of assistance occurring for each support point based on the history of past remote assistance, and records this in the operator information database 74. The recording unit 68 also records the content of assistance in the operator information database 74. The content of assistance includes, for example, right turn assistance or left turn assistance at intersections, door opening / closing assistance at bus stops, and starting assistance.

[0031] The allocation unit 69 allocates operators to the assistance points included in the route of each vehicle 10 based on the start timing of the remote assistance predicted by the prediction unit 66.

[0032] 2 is a diagram showing the hardware configuration of the control unit 60. The control unit 60 is configured as a device including a general computer.

[0033] 2, the control unit 60 includes a central processing unit (CPU) 60A, a read-only memory (ROM) 60B, a random access memory (RAM) 60C, a nonvolatile memory 60D, and an input / output interface (I / O) 60E. The CPU 60A, the ROM 60B, the RAM 60C, the nonvolatile memory 60D, and the I / O 60E are connected to each other via a bus 60F. The communication unit 52 and the storage device 70 are connected to the I / O 60E.

[0034] CPU 60A is an example of a computer. The term "computer" here refers to a processor in a broad sense, and includes general-purpose processors (e.g., CPUs) and dedicated processors (e.g., GPUs: Graphics Processing Units, ASICs: Application Specific Integrated Circuits, FPGAs: Field Programmable Gate Arrays, programmable logic devices, etc.).

[0035] The storage device 70 is configured with a non-volatile storage device such as a hard disk. The CPU 60A reads and executes a remote assistance program 76 stored in the storage device 70. As a result, the CPU 60A functions as each functional unit constituting the control unit 60 described above. The remote assistance program 76 may be realized by being stored in a non-volatile non-transitory recording medium or distributed via a network and installed appropriately in the remote assistance device 50.

[0036] Examples of non-volatile non-transitory recording media include CD-ROMs (Compact Disc Read Only Memory), magneto-optical disks, HDDs (Hard Disk Drives), DVD-ROMs (Digital Versatile Disc Read Only Memory), flash memory, memory cards, etc.

[0037] The operator terminal 80 is configured as a computer or mobile terminal equipped with a CPU and memory. The autonomous driving system 100 in this embodiment includes one or more operator terminals 80. Each operator terminal 80 includes an operation unit 82 and a display unit 84. The operation unit 82 accepts operations from the operator to remotely support the vehicle 10. The accepted operations are transmitted as operation signals to the vehicle 10 via the remote support device 50. The display unit 84 displays a user interface used for operation by the operator. The display unit 84 also displays images from the camera 14 transmitted from the vehicle 10.

[0038] The remote assistance process executed by the control unit 60 of the autonomous driving system 100 will be described with reference to Fig. 3. The remote assistance process shown in Fig. 3 is a process that is repeatedly executed at a predetermined interval by the remote assistance device 50 while the autonomous driving system 100 is operating. Note that the following description will be given assuming that the route of each vehicle 10 has been specified by the route planning unit 62.

[0039] In step S100, it is determined whether vehicle information has been received from vehicle 10. As described above, vehicle information is periodically transmitted from vehicle 10, and includes identification information (vehicle ID) and location information of vehicle 10.

[0040] If the vehicle information has not been received, this routine is terminated, and if the vehicle information has been received, the routine proceeds to step S102.

[0041] In step S102, an assistance timing prediction process is executed. Here, the assistance timing prediction process is a process of predicting the timing at which the operator will start remote assistance based on the route specified by the route planning unit 62 and the vehicle information received from the vehicle 10.

[0042] Once the routes of the vehicles 10 are identified, the estimated time of arrival of each vehicle 10 at the support point is identified. Furthermore, when vehicle information is received from the vehicle 10, the actual position of the vehicle 10 at the current time is determined. Therefore, in step S102, the arrival time at each support point is updated based on the real-time position information of the vehicle 10.

[0043] Furthermore, in step S102, the predicted result of the timing at which the operator will start remote support is updated based on the updated arrival time. In this embodiment, the timing at which the vehicle 10 arrives at the support point is predicted as the timing at which the operator will start remote support. Note that the timing at which the operator will start remote support does not have to be the same as the timing at which the vehicle 10 arrives at the support point, and may be, for example, earlier or later than the timing at which the vehicle 10 arrives at the support point.

[0044] In step S104, a process for predicting the number of operators required is executed. This process for predicting the number of operators required for each time period from the present to the future.

[0045] In step S104, the number of vehicles that will receive remote support during the same time period can be determined from the estimated time each vehicle will arrive at the support point and the required time at the support point, and this number of vehicles is then predicted as the required number of operators. However, a margin may be added to the predicted number of operators in advance to allow for prediction errors and operator breaks.

[0046] Before starting work, the administrator can roughly determine the number of operators required for each time period of the day based on the history of past operator numbers, such as the same time period in the past or the same time period on the same day of the week.

[0047] In step S106, an operator reservation process is executed. In the operator reservation process, a determination is made as to whether or not a reservation for operator support is necessary based on the support occurrence probability calculated for each support point based on the history of past remote support and the priority set according to the support content, which indicates the importance when support is needed, and a reservation for support is set if support is necessary. For example, if the priority is equal to or higher than a threshold, it is determined that a reservation is necessary. On the other hand, if the priority is less than the threshold, a reservation necessity determination table indicating the correspondence between the support occurrence probability, priority, and whether or not a reservation is necessary is referenced to determine whether or not a reservation is necessary. Note that the method for determining whether or not a reservation for support is necessary is not limited to this.

[0048] In step S107, it is determined whether or not there is a prediction that there will be no available operators. A prediction that there will be no available operators means that there is a schedule in which the number of available operators will be less than the number of vehicles 10 determined to require remote assistance in step S106.

[0049] If it is predicted that there will be no more available operators, the process proceeds to step S112. On the other hand, if it is predicted that there will be available operators, the process proceeds to step S108.

[0050] In step S108, based on the vehicle information received from each vehicle 10, it is determined whether or not there is a vehicle 10 approaching the support point or whether or not a support request has been received from a vehicle 10. For example, the remote support device 50 determines that the vehicle 10 is approaching the support point when it predicts that the vehicle 10 will arrive at the next support point on the route of the vehicle 10 within a predetermined time (for example, 10 seconds).

[0051] If there is a vehicle 10 approaching the support point or if a support request has been received from the vehicle 10, the process proceeds to step S109, and if there is no vehicle 10 approaching the support point and no support request has been received from the vehicle 10, the routine is terminated.

[0052] In step S109, an operator allocation process is executed, which is a process of allocating an operator to a vehicle 10 that requires remote assistance.

[0053] In step S110, it is determined whether there are currently no available operators. When there are currently no available operators, this means that the number of available operators at present is less than the number of vehicles 10 that require remote support, and there are vehicles 10 that cannot receive remote support.

[0054] If there are no available operators at this time, the process proceeds to step S112. On the other hand, if there are available operators, the process proceeds to step S114.

[0055] In step S112, a process is executed to deal with the case where there is no available operator or it is predicted that there will be no available operator, as will be described in detail later.

[0056] In step S114, remote assistance processing is executed. As soon as the vehicle 10 arrives at the assistance point or as soon as an assistance request is received from the vehicle 10, the operator uses the operator terminal 80 to perform operations to remotely assist the vehicle 10. Therefore, in step S114, an operation signal is received from the operator terminal 80 and the operation signal is transferred to the vehicle 10, thereby performing remote assistance. Note that the operation signal for remote assistance may be exchanged directly between the vehicle 10 and the operator terminal 80 without going through the remote assistance device 50.

[0057] In step S116, a database update process is executed. In the database update process, the time required from the start to the end of the remote support performed in step S114 and the number of operators who provided the remote support are recorded in the operator information database 74, and the operator information database 74 is updated.

[0058] A specific example of the response process will be described below.

[0059] As a countermeasure, the control unit 60 controls the operation of the vehicle 10 so that there are no vacant operators at the support point.

[0060] For example, if the number of available operators falls below the number of remote assistance requests scheduled to occur simultaneously due to reasons such as sudden assistance occurring or changes in route settings or operation schedules, vehicles to make up for the shortage of operators are selected, and vehicle instructions such as stopping instructions, waiting at a bus stop, driving at a slow speed, and route changes are given to the selected vehicles.

[0061] Furthermore, among the vehicles for which remote assistance is predicted, vehicle instructions are given to the vehicles with a shortage of operators, starting with the vehicle that will have the least impact on the feasibility of the autonomous driving service. However, if there are multiple vehicles with the same impact on the feasibility of the service, the vehicle with the worst prediction accuracy for the assistance period, i.e., the vehicle farthest from the assistance point, is selected, and if the impact on the feasibility of the service does not change depending on the timing of the instruction, instructions are given when the prediction accuracy becomes higher, i.e., when the vehicle approaches the assistance point. This makes it possible to prevent ineffective instructions from being given due to poor prediction accuracy.

[0062] In addition, among the vehicles predicted to require assistance, instructions are given to the vehicles lacking operators in order of the vehicle that will have the least impact on safety and the feasibility of autonomous driving services. However, if there is a vehicle with a low probability of requiring assistance, a vehicle that can handle the situation will be selected after it is determined whether assistance will be required for that vehicle. This prevents ineffective vehicle instructions due to a low probability of assistance.

[0063] Incidentally, the longer the time until remote assistance is disclosed, the more likely it is that there will be more options for giving instructions to the vehicle. In addition, it is generally believed that instructions that can be given in advance, such as route changes, will have a smaller impact on the feasibility of automated driving services, such as the impact of congestion on surrounding roads and bus schedules, than instructions that require an emergency stop of the vehicle when there is no operator available.

[0064] Therefore, when there are no available operators or when it is predicted that there will be no available operators, and vehicle instructions for the vehicle can be selected from multiple options, instructions may be given to the vehicle according to the length of time until remote assistance begins.

[0065] For example, if the period until remote assistance begins is relatively short, it is generally better to issue a first instruction, such as to stop the vehicle, but if the period until remote assistance begins is relatively long, it is better to issue a second instruction, such as to change the route. Therefore, if there is no available bus for the operator or it is predicted that there will be no available bus, the second instruction to drive the vehicle to a bus stop and stop there may be prioritized over the first instruction to make an emergency stop of the vehicle. Also, the second instruction to drive the vehicle at a slow speed may be prioritized over the first instruction to make an emergency stop of the vehicle. Also, the second instruction to change the route may be prioritized over the first instruction to make an emergency stop of the vehicle. Also, the second instruction to change the route may be prioritized over the first instruction to evacuate the vehicle to a safety zone. Also, the second instruction to change the route may be prioritized over the first instruction to stop the vehicle at a bus stop.

[0066] Furthermore, when there are no available operators or when it is predicted that there will be no available operators, and vehicle instructions for a vehicle can be selected from multiple options, the vehicle instruction may be selected according to the impact on the feasibility of autonomous driving services. For example, the instruction with the least impact on autonomous driving services may be selected depending on the vehicle. For example, if the vehicle is a bus, the first instruction to stop at a bus stop may be prioritized over the second instruction to change the route. This is because changing the route in a bus would have a significant impact on the bus's punctuality and surrounding traffic. Even when the first instruction is selected in this way, the second instruction may ultimately have a smaller impact if a situation changes, such as an additional unexpected assistance request. Therefore, the second instruction may be selected if the probability of assistance occurring is high or the probability of assistance occurring is unknown. For example, by issuing the second instruction to change the route to vehicle A instead of the first instruction to stop at a bus stop, remote assistance is unnecessary, and it is possible to prevent vehicle A from waiting at the bus stop for a long period of time or vehicle B from being instructed to make an emergency stop if unexpected assistance is required from vehicle B.

[0067] In addition, instructions may be selected that minimize the impact on the entire service, including multiple vehicles. This is because the impact may not be minimized if instructions to individual vehicles are considered alone. For example, if there are no available operators or it is predicted that there will be no available operators, the impact on each vehicle will be smaller if both vehicle A and vehicle B are instructed to stop at a bus stop rather than to change their route. However, if the support periods overlap, either vehicle A or vehicle B may be instructed to change their route or to travel at a slower speed. This eliminates the overlap of the support periods and shortens waiting times at bus stops, resulting in improved punctuality and minimizing the impact on the entire service.

[0068] In principle, instructions to the vehicle are given based on a prediction of the most recent occurrence of assistance so that instructions to the vehicle do not need to be updated, but it is also possible to monitor the vehicle's position and update the instructions to the vehicle if the position deviates from the prediction of the occurrence of assistance.

[0069] A specific example will be described below.

[0070] (Example 1)

[0071] In specific example 1, as shown in Fig. 4, it is assumed that different routes are specified for vehicles A, B, and C by the route planning unit 62 of the remote support device 50. In the example of Fig. 4, a route RA is set for vehicle A. A route RB that passes through bus stop BS is set for vehicle B. A route RC is set for vehicle C.

[0072] In such a case, if there are no available operators, the control unit 60 instructs vehicles A, B, and C to stop, and if there are available operators, it instructs vehicles A, B, and C to resume traveling. Note that stopping does not only mean stopping immediately at the location, but also includes driving to a safety zone and then stopping. Here, a safety zone refers to a place that does not interfere with the traveling of other vehicles, such as a road shoulder, a bus stop, or a parking lot.

[0073] Figure 5 shows the time series of the driving conditions when instructions are given to each vehicle and when they are not given. The horizontal axis represents time.

[0074] As shown in Fig. 5, the control unit 60 instructs vehicles A, B, and C to stop when there are no available operators (0). As a result, as shown in Fig. 4, each vehicle stops on the shoulder of the road, which serves as a safety zone, as indicated by the solid arrows. Note that as shown in Fig. 4, vehicle A is in the state of starting to enter the intersection, and in order to avoid stopping at the intersection, the control unit 60 instructs vehicle A to drive to the safety zone beyond the intersection and then stop.

[0075] Then, as shown in Figure 5, when there are available operators, that is, when there is one or more available operators, the system instructs vehicles A, B, and C to resume traveling. As a result, vehicles A, B, and C resume traveling. Note that vehicle B has bus stop BS on route RB, so it stops at the bus stop and then departs from the bus stop.

[0076] In this way, when there are no available operators, all vehicles are stopped, and when there is one or more available operators, they are instructed to resume driving. This makes it possible to shorten the time spent waiting for assistance at a support point if there is one on the route of each vehicle.

[0077] (Example 2)

[0078] In Example 2, if there is no support point on the route and no remote support is required, no instruction to stop, etc. is given to a vehicle. In addition, if there is a support point on the route, the vehicle is instructed to stop.

[0079] In Example 2, as shown in Fig. 6, a route RA that does not pass through any support points is set for vehicle A. A route RB that passes through bus stop BS and support point SP1 is set for vehicle B. A route RC that passes through support point SP2 is set for vehicle C.

[0080] As shown in Figure 7, if there is no instruction to stop the vehicle, vehicle A will continue on route RA without stopping, as there is no support point on route RA. Vehicle B has bus stop BS and support point SP1 on route RB, so it stops at bus stop BS before departing, and receives remote support from an operator after arriving at support point SP1. Vehicle C also has support point SP2 on route RC, so it receives remote support from an operator after arriving at support point SP2. In Figure 7, the width of the rectangular area labeled "support point" with a hatched background represents the length of time required for remote support at the support point. As shown in Figure 7, the periods during which vehicles B and C receive remote support overlap, so they will have to wait for support at the support point.

[0081] Therefore, as shown in Fig. 7, the remote support device 50 instructs vehicle B to stop immediately after arriving at bus stop BS, which is a safety zone. It also instructs vehicle C to stop immediately. Note that vehicle A does not need to be instructed to stop, etc., because there is no support point on route RA. Therefore, no particular instruction to stop, etc. is given to vehicle A.

[0082] As a result, as shown by the solid arrow in Figure 6, vehicle B stops after arriving at the bus stop. Vehicle C also stops immediately. This makes it possible to avoid overlapping periods during which vehicles B and C receive remote assistance. Then, when there are available operators, vehicle B is instructed to resume traveling. This allows vehicle B to depart bus stop BS and immediately receive remote assistance from an operator at assistance point SP2 without having to wait for assistance. Note that it is preferable to avoid overlapping periods during which remote assistance is received, but it is not necessary to completely avoid overlapping as long as this reduces the overlapping periods during which remote assistance is received. The same applies to subsequent overlapping periods during which remote assistance is received.

[0083] (Example 3)

[0084] In specific example 3, if there is no support point on the route and no remote support is required, no instruction is given to stop, etc. Furthermore, among vehicles that have a support point on the route, those that can change to another route that can avoid the support point are instructed to change their route, and those that cannot change their route are instructed to stop.

[0085] In the specific example 3, as shown in FIG. 8, routes RA, RB, and RC similar to those in FIG. 6 are set for vehicles A, B, and C.

[0086] As shown in Figure 9, when there is no instruction to stop the vehicle, the situation is the same as in Figure 7, and therefore a description thereof will be omitted. Also, what makes Specific Example 3 different from Specific Example 2 is the instruction given to vehicle C, while the instructions given to vehicles A and B are the same as in Specific Example 2, and therefore a description thereof will be omitted.

[0087] As shown in FIG. 8, another route RC2 that can avoid the support point SP2 can be set for the vehicle C, so the vehicle C is instructed to change from the route RC to the route RC2.

[0088] As a result, as shown in Figure 9, vehicle B stops after arriving at the bus stop. Also, vehicle C travels while avoiding support point SP2. This makes it possible to prevent vehicles B and C from receiving remote support during overlapping periods. Then, when there are available operators, vehicle B is instructed to resume traveling. This allows vehicle B to depart bus stop BS and immediately receive remote support from an operator without having to wait for support at support point SP1.

[0089] (Example 4)

[0090] In the fourth specific example, the same routes RA, RB, and RC as those in FIG. 8 are set.

[0091] As shown in Figure 10, when there is no instruction to stop the vehicle, the situation is the same as in Figure 9, and therefore a description thereof will be omitted. Also, what makes Specific Example 4 different from Specific Example 3 is the instruction given to vehicle B, while the instructions given to vehicles A and C are the same as in Specific Example 3, and therefore a description thereof will be omitted.

[0092] As shown in Figure 10, when the number of available operators reaches zero, vehicle B is not instructed to stop, but rather to delete the route from bus stop BS onwards from route RB. As a result, vehicle B will wait at bus stop BS when it arrives there. Then, when the number of available operators reaches one or more, the route from bus stop BS onwards is reset, i.e., restored. This allows vehicle B to resume traveling, and it can receive remote assistance from an operator at support point SP1 without having to wait for assistance.

[0093] (Example 5)

[0094] In the fifth specific example, when it is predicted that there will be no available operators, an instruction to stop or change the route is given so that the overlap of the remote support periods of each vehicle is reduced.

[0095] In Example 5, as shown in Fig. 11, a route RA that passes through support point SP1 is set for vehicle A. A route RB that passes through bus stop BS and support point SP2 is set for vehicle B. A route RC that passes through support point SP3 is set for vehicle C.

[0096] In Example 5, we will explain a situation in which it is predicted that remote support will be required for vehicle B, and therefore one operator is scheduled to be assigned, but it is newly discovered that remote support will also be required for vehicles A and C, resulting in a prediction that there will be a shortage of two operators.

[0097] As shown in Figure 12, if there is no instruction to stop the vehicle, for example, vehicle A has support point SP1 on route RA, so a support scheduled task occurs when route RA is set. For vehicle B, there is support point SP2 on route RB, so a support scheduled task occurs when route RB is set. For vehicle C, there is support point SP3 on route RC, so a support scheduled task occurs when route RC is set. In this way, a support scheduled task occurs for each of vehicles A to C, but because one operator is scheduled to be assigned, one of the three vehicles A to C will not wait for support, but two will wait for support due to a shortage of operators.

[0098] Therefore, as shown in Fig. 12, the control unit 60 does not instruct vehicle A to stop or the like, but allows it to continue traveling. It instructs vehicle B to wait at bus stop BS. It also instructs vehicle C to change its route to route RC2, which passes through support point SP4, which is earlier than support point SP3, as shown in Fig. 11.

[0099] As a result, vehicle C will provide remote support earlier than if it had traveled along the original route RC. Furthermore, vehicle B will wait at bus stop BS, so the timing of remote support will be delayed. By issuing instructions to vehicles B and C in this manner, it is possible to avoid overlapping of the remote support periods of vehicles A, B, and C, as shown in FIG. 12. Therefore, even if it is predicted that there will be a shortage of two operators when routes RA, RB, and RC are initially set, it is possible to avoid a shortage of operators and avoid waiting for support at support points.

[0100] It is not appropriate to change the route if vehicle C is a vehicle that cannot change its route, such as a bus. Therefore, if vehicle C is a vehicle that cannot change its route, such as a bus, it may be instructed to travel at a speed lower than the set speed so that vehicle C arrives at support point SP3 after the remote support of vehicle B has finished, rather than instructing vehicle C to change its route.

[0101] (Example 6)

[0102] In the sixth specific example, a case will be described in which instructions to a vehicle are controlled based on a support occurrence probability that indicates the probability of support occurring at a support point.

[0103] In specific example 6, for example, as shown in the timing chart in the upper left of Fig. 13, when there are no instructions to the vehicles, for vehicle A, a route RA is set as in Fig. 11, a support scheduled task occurs, and the support occurrence probability is 30%. Also, for vehicle B, a route RB is set as in Fig. 11, a support scheduled task occurs, and the support occurrence probability is 100%. Also, for vehicle C, a route RC is set as in Fig. 11, a support scheduled task occurs, and the support occurrence probability is 20%. Also, for vehicle B, an operator is assigned and support is scheduled, but for vehicles A and C, operators cannot be assigned, and there is a shortage of two operators.

[0104] In this case, if the probability of assistance occurring is equal to or less than a predetermined threshold (e.g., 30%), the control unit 60 withholds instructions until it is determined whether assistance will occur at the assistance point. Therefore, for vehicles A and C, the probability of assistance occurring is 30% or less, and there is a high possibility that remote assistance will not occur, so instructions are withheld for vehicles A and C until the vehicles notify them of whether assistance will occur.

[0105] In this case, as shown in the timing chart in the upper right of Figure 13, assume that remote assistance is not actually provided to vehicle A alone. In this case, remote assistance is provided to vehicle C before vehicle B, but this overlaps with the remote assistance period for vehicle B. Therefore, vehicle B is instructed to wait at bus stop BS. Then, after the remote assistance for vehicle C has finished, vehicle B is instructed to resume driving. This makes it possible to avoid overlapping of the remote assistance periods for vehicle B and vehicle C, and to avoid a shortage of operators. Furthermore, since instructions are withheld for vehicle A until it is determined that there will be no remote assistance, it is possible to avoid giving unnecessary instructions.

[0106] Also, as shown in the timing chart at the bottom right of Figure 13, assume that remote assistance is not actually provided to vehicle C alone. In this case, remote assistance is provided to vehicle A before vehicle B, but this overlaps with the remote assistance period for vehicle B. Therefore, vehicle B is instructed to wait at bus stop BS. Then, after the remote assistance for vehicle A ends, vehicle B is instructed to resume driving. This makes it possible to avoid overlapping of the remote assistance periods for vehicle A and vehicle C, and to avoid a shortage of operators. Furthermore, since instructions to vehicle C are withheld until it is determined that no remote assistance will be provided, it is possible to avoid giving unnecessary instructions.

[0107] Also, as shown in the timing chart at the bottom left of Figure 13, assume that remote assistance is not actually provided to either vehicle A or C. In this case, vehicle B arrives at bus stop BS after it is determined that remote assistance will not be provided by vehicle A. At that point, it is not yet certain whether remote assistance will be provided by vehicle C, so there is a possibility that the remote assistance periods for vehicles B and C may overlap. Therefore, vehicle B is instructed to wait at bus stop BS. After that, when it is determined that remote assistance will not be provided by vehicle C, vehicle B is instructed to resume traveling. Furthermore, since instructions are withheld for vehicles A and C until it is determined that remote assistance will not be provided, it is possible to avoid issuing unnecessary instructions.

[0108] The remote assistance device 50 can provide remote assistance to each vehicle 10 by repeatedly executing the remote assistance process described above.

[0109] According to the remote assistance device 50 of the present embodiment described above, when there are no operators available to provide remote assistance or when it is predicted that there will be no more available operators, the operation of the vehicle 10 is controlled so that there will be no more available operators at the assistance point. This reduces the time spent waiting for assistance at the assistance point. This improves safety and enables smooth operation with remote assistance.

[0110] The present disclosure is not limited to the above-described embodiments, and various modifications and applications are possible without departing from the spirit and scope of the present invention.

[0111] In addition, the configuration of the autonomous driving system 100 described in the above embodiment (see Figure 1) is one example, and it goes without saying that unnecessary parts may be deleted or new parts may be added within the scope of the present invention.

[0112] Furthermore, the processing flow of the remote assistance program 76 described in the above embodiment (see FIG. 3) is also an example, and it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged within the scope of the present invention. [Explanation of symbols]

[0113] 10 vehicle, 50 remote assistance device, 60 control unit, 62 route planning unit, 64 remote assistance unit, 66 prediction unit, 68 recording unit, 69 allocation unit, 70 storage device, 100 automated driving system

Claims

1. A control unit (60) that controls the operation of the autonomous vehicle (10) so that there are no available operators at the support point where remote support is provided to the autonomous vehicle (10) when there are no available operators to provide the remote support or when it is predicted that there will be no available operators at the support point. Equipped with The control unit instructs the autonomous vehicle to stop when the operator is unavailable, and instructs the autonomous vehicle to resume traveling when the operator is available; the control unit instructs the autonomously driven vehicle to stop when the assistance point is present on a route along which the autonomously driven vehicle is traveling; The control unit instructs the autonomous vehicle to stop by instructing the deletion of the route, and when an available position becomes available for the operator, instructs the autonomous vehicle to restart traveling by instructing the autonomous vehicle to reset the route. A remote support device (50).

2. When a safety zone exists on the route, the control unit instructs the vehicle to drive to the safety zone and then stop. The remote assistance device according to claim 1.

3. When there is another route that can avoid the assistance point, the control unit instructs the autonomously driven vehicle to change to the other route. The remote support device according to claim 1 or 2.

4. A control unit (60) that controls the operation of an autonomous vehicle (10) so that an operator is not unavailable at a support point where remote support is provided to the autonomous vehicle (10) when the operator is unavailable or when it is predicted that the operator will be unavailable at the support point. Equipped with When it is predicted that there will be no more available operators, the control unit controls operation of at least one of the plurality of autonomous vehicles so as to reduce overlap of remote assistance periods required for remote assistance of the plurality of autonomous vehicles. Remote assistance devices.

5. The control unit instructs the at least one autonomously driven vehicle to stop. The remote assistance device according to claim 4.

6. When there is another route that can avoid the assistance point for the at least one of the autonomous vehicles, the control unit instructs the at least one of the autonomous vehicles to change to the other route.

6. The remote support device according to claim 4 or claim 5.

7. The control unit controls the driving of the autonomous vehicle based on the assistance occurrence probability at the assistance point. The remote support device according to any one of claims 4 to 6.

8. The control unit withholds instructions until it is determined whether or not support will be provided at the support point. The remote assistance device according to claim 7.

9. The control unit instructs the at least one autonomously driven vehicle to travel at a speed lower than a set speed. The remote support device according to any one of claims 4 to 8.

10. When the number of available operators is less than the number of remote support requests scheduled to occur simultaneously, the control unit selects vehicles to accommodate the shortage of operators and issues vehicle instructions. The remote support device according to any one of claims 1 to 9.

11. The control unit issues vehicle instructions to vehicles for which remote assistance is predicted to be required, in order of the number of vehicles for which there is a shortage of operators, starting with the vehicle that will have the least impact on the feasibility of the autonomous driving service. The remote support device according to any one of claims 1 to 10.

12. A control unit (60) that controls the operation of an autonomous vehicle (10) so that an operator is not unavailable at a support point where remote support is provided to the autonomous vehicle (10) when the operator is not available or when it is predicted that the operator will not be available at the support point. Equipped with When there is no available operator or when it is predicted that there will be no available operator, and when a vehicle instruction for the vehicle can be selected from a plurality of options, the control unit performs a vehicle instruction according to the length of the period until remote assistance is started. Remote assistance devices.

13. When there is no available operator or when it is predicted that there will be no available operator, and when a vehicle instruction to a vehicle can be selected from a plurality of options, the control unit issues a vehicle instruction according to the impact on the feasibility of a service related to autonomous driving. A remote assistance device according to any one of claims 1 to 12.

14. The control unit issues vehicle instructions that minimize the impact on the entire service including multiple vehicles. A remote assistance device according to any one of claims 1 to 13.

15. A computer (60A) When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; If there is no available operator, instruct the autonomous vehicle to stop, and if there is available operator, instruct the autonomous vehicle to resume traveling; instructing the autonomous vehicle to stop when the assistance point is present on the route traveled by the autonomous vehicle; The system instructs the automated driving vehicle to stop by instructing the deletion of the route, and when an available space becomes available for the operator, the system instructs the automated driving vehicle to restart traveling by instructing the automated driving vehicle to reset the route. Remote assistance methods.

16. A computer (60A), When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; When it is predicted that there will be no vacant operators, the operation of at least one of the plurality of autonomous vehicles is controlled so as to reduce overlap of remote assistance periods required for remote assistance of the plurality of autonomous vehicles. Remote assistance methods.

17. A computer (60A), When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; When there are no operators available or when it is predicted that there will be no operators available, and when vehicle instructions for a vehicle can be selected from multiple options, vehicle instructions are given according to the length of the period until remote support begins. Remote assistance methods.

18. On the computer, When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; If there is no available operator, instruct the autonomous vehicle to stop, and if there is available operator, instruct the autonomous vehicle to resume traveling; instructing the autonomous vehicle to stop when the assistance point is present on the route traveled by the autonomous vehicle; The system instructs the automated driving vehicle to stop by instructing the deletion of the route, and when an available space becomes available for the operator, the system instructs the automated driving vehicle to restart traveling by instructing the automated driving vehicle to reset the route. A remote support program (76) for executing a process including the above.

19. On the computer, When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; When it is predicted that there will be no vacant operators, the operation of at least one of the plurality of autonomous vehicles is controlled so as to reduce overlap of remote assistance periods required for remote assistance of the plurality of autonomous vehicles. A remote support program (76) for executing a process including the above.

20. On the computer, When there is no availability of an operator to provide remote assistance at an assistance point where remote assistance is provided to an autonomous vehicle, or when it is predicted that there will be no availability of an operator to provide the remote assistance, the autonomous vehicle is controlled to prevent the availability of the operator at the assistance point from becoming full; When there are no operators available or when it is predicted that there will be no operators available, and when vehicle instructions for a vehicle can be selected from multiple options, vehicle instructions are given according to the length of the period until remote support begins. A remote support program (76) for executing a process including the above.

Images