Remote support system and remote support method

Abstract

To reduce a load to be placed on an operator who remotely supports travel of an automatic driving vehicle.SOLUTION: A remote support system comprises a result database. The result database shows results of support request processing by an automatic driving vehicle for every scene. The remote support system requests an instruction to an object automatic driving vehicle to an operator in response to a support request from the object automatic driving vehicle. Support processing is vehicle processing to be performed by the operator instead of the object automatic driving vehicle. On the other hand, commission processing is processing to be committed to the object automatic driving vehicle by the operator. The remote support system determines recommended options of combination of the support processing and the commission processing on the basis of the result database. Then, the remote support system presents the recommended options to the operator.SELECTED DRAWING: Figure 3

Description

[Technical field]

[0001] The present invention relates to a technology for remotely supporting the driving of an autonomous vehicle. [Background technology]

[0002] Patent Document 1 discloses a control device that remotely controls an autonomous vehicle. After remote control of a target vehicle is performed, the control device registers location data including position information and sensing data of the target vehicle. When performing remote control, the control device determines whether registered location data similar to the current location data acquired from the target vehicle is available. If the similar registered location data is available, the control device performs a first remote control using the similar registered location data and without operator intervention. On the other hand, if the similar registered location data is not available, the control device performs a second remote control with operator intervention.

[0003] Patent Document 2 discloses a management device that manages vehicles. The vehicle uses sensors to create surrounding information that indicates the surrounding situation, and performs autonomous driving based on the surrounding information. In a situation where autonomous driving is difficult, the management device acquires surrounding information from the vehicle, and selects either remote control of the vehicle or alternative processing according to the surrounding information. The alternative processing is changing the sensor used or changing the autonomous driving route.

[0004] Patent Document 3 discloses a vehicle remote operation support system. The vehicle remote operation support system includes a vehicle that performs autonomous driving and a control center having an operator. When autonomous driving is difficult, the vehicle transmits surrounding information to the control center and calls the operator. The operator inputs parameters for selecting a driving pattern for remotely controlling the vehicle according to the surrounding information. The control center selects a driving pattern based on the input parameters, and generates a route for the vehicle using the selected driving pattern and the surrounding information. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] JP 2019-185280 A [Patent Document 2] JP 2019-207539 A [Patent Document 3] JP 2019-191982 A Summary of the Invention [Problem to be solved by the invention]

[0006] In response to an assistance request from an autonomous vehicle, an operator may remotely assist the autonomous vehicle in traveling. In this case, if the burden on the operator increases, the efficiency of the remote assistance may decrease.

[0007] One object of the present invention is to provide a technique that can reduce the burden on an operator who remotely assists the driving of an autonomous vehicle. [Means for solving the problem]

[0008] The first aspect relates to a remote assistance system that remotely supports the driving of an autonomous vehicle. Self-driving vehicles are A plurality of vehicle processes including recognition and decision making during autonomous driving; an assistance request process for sending an assistance request to a remote assistance system, the assistance request requesting assistance for at least one of the plurality of vehicle processes; Execute. The remote support system is one or more processors; A performance database showing the performance of autonomous vehicles in handling assistance requests for each scene. Equipped with. The one or more processors an operator instruction request process that requests an operator to provide instructions to the target autonomous vehicle in response to an assistance request sent from the target autonomous vehicle in the first scene; An operator instruction notification process that notifies the target autonomous vehicle of instructions from the operator; The apparatus is configured to execute the following steps: The assistance process is one of the multiple vehicle processes in the first scene that is performed by an operator on behalf of the target autonomous vehicle. The delegated process is a vehicle process among the multiple vehicle processes in the first scene that the operator entrusts to the target autonomous vehicle. In the operator instruction request processing, the one or more processors Retrieving a track record of assistance request processing related to the first scene from a track record database; determining a recommended option to be recommended to the operator from among a plurality of options of combinations of assistance processing and delegation processing based on a track record related to the first scene; At least recommended options are presented to the operator.

[0009] The second aspect relates to a remote assistance method for remotely assisting the driving of an autonomous vehicle. Self-driving vehicles are A plurality of vehicle processes including recognition and decision making during autonomous driving; an assistance request process for sending an assistance request to a remote assistance system, the assistance request requesting assistance for at least one of the plurality of vehicle processes; Execute. The performance database indicates the performance of assistance request processing by autonomous vehicles for each scene. Remote support methods include: an operator instruction request process that requests an operator to provide instructions to the target autonomous vehicle in response to an assistance request sent from the target autonomous vehicle in the first scene; An operator instruction notification process that notifies the target autonomous vehicle of instructions from the operator; Includes. The assistance process is one of the multiple vehicle processes in the first scene that is performed by an operator on behalf of the target autonomous vehicle. The delegated process is a vehicle process among the multiple vehicle processes in the first scene that the operator entrusts to the target autonomous vehicle. The operator instruction request process is obtaining a performance record of the assistance request process related to the first scene from a performance database; determining a recommended option to be recommended to an operator from among a plurality of options of combinations of assistance processing and delegation processing based on a performance of assistance request processing related to the first scene; A process of presenting at least recommended options to an operator; Includes. Effect of the Invention

[0010] According to the present invention, in response to an assistance request from a target autonomous vehicle, an operator is requested to give instructions to the target autonomous vehicle. Assistance processing is vehicle processing performed by the operator on behalf of the target autonomous vehicle. On the other hand, delegated processing is processing that the operator leaves to the target autonomous vehicle. As the number of delegated processing increases and the number of assistance processing decreases, the number of matters that the operator must decide decreases accordingly. This contributes to reducing the burden on the operator.

[0011] Furthermore, according to the present invention, recommended options for combinations of assistance processes and delegated processes are determined based on a performance database showing the performance of assistance request processes. The recommended options are then presented to the operator. By referring to the recommended options, the operator can reduce the time it takes to consider which vehicle process to delegate. This reduces the burden on the operator.

[0012] The burden on the operator is reduced, and the time it takes the operator to handle a single assistance request (i.e., a single target autonomous vehicle) is shortened. As a result, the efficiency of remote assistance is improved. This also means that a limited number of operators can handle more assistance requests (target autonomous vehicles). This is preferable from a cost perspective. According to the present invention, it is possible to improve the efficiency of remote assistance while suppressing increases in costs. [Brief description of the drawings]

[0013] [Figure 1]FIG. 1 is a conceptual diagram for explaining an overview of remote support according to an embodiment of the present invention. [Diagram 2] 1 is a conceptual diagram for explaining an overview of a remote support system according to an embodiment of the present invention. [Diagram 3] FIG. 2 is a conceptual diagram for explaining a remote support process by the remote support system according to the embodiment of the present invention. [Figure 4] 11 is a conceptual diagram for explaining an example of an operator instruction request process by the remote support system according to the embodiment of the present invention. FIG. [Diagram 5] 1 is a block diagram showing an example of the configuration of an autonomous driving system mounted on an autonomous driving vehicle according to an embodiment of the present invention. FIG. [Figure 6] 2 is a block diagram showing an example of driving environment information according to an embodiment of the present invention; [Figure 7] 1 is a block diagram showing an example of a configuration of a remote support system according to an embodiment of the present invention. [Figure 8] FIG. 2 is a conceptual diagram illustrating an example of a performance database according to an embodiment of the present invention. [Figure 9] FIG. 13 is a conceptual diagram showing another example of the performance database according to the embodiment of the present invention. [Figure 10] 5 is a block diagram showing an example of various information related to a remote assistance process according to an embodiment of the present invention. FIG. [Figure 11] 10 is a flowchart showing an operator instruction request process in the remote support process according to the embodiment of the present invention. [Figure 12] 10 is a flowchart showing an operator instruction notification process in the remote assistance process according to the embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] An embodiment of the present invention will be described with reference to the accompanying drawings.

[0015] 1. Overview of remote support FIG. 1 is a conceptual diagram for explaining an overview of remote assistance according to the present embodiment. The autonomous vehicle 1 is a vehicle capable of autonomous driving. The autonomous driving here is assumed to be based on the premise that the driver does not necessarily have to concentrate 100% on driving (so-called autonomous driving at level 3 or higher). The autonomous vehicle 1 may be an autonomous vehicle at level 4 or higher that does not require a driver. An example of such an autonomous vehicle 1 is an autonomous bus that operates within a limited area.

[0016] The remote assistance system 100 is installed, for example, in an autonomous driving center in an area where a large number of autonomous vehicles 1 travel. The autonomous vehicles 1 and the remote assistance system 100 are capable of communicating with each other. The remote assistance system 100 communicates with the autonomous vehicles 1 and remotely assists the traveling of the autonomous vehicles 1. More specifically, a human operator remotely assists the traveling of the autonomous vehicles 1 via the remote assistance system 100. It can also be said that the remote assistance system 100 is a system that assists the operator in remotely assisting the autonomous vehicles 1.

[0017] For example, as shown in Fig. 2, remote assistance system 100 includes a monitor. Information such as the situation of the area, the position and status of each autonomously driven vehicle 1, etc. is displayed on the monitor. An operator monitors the information displayed on the monitor. Then, the operator remotely assists the traveling of autonomously driven vehicle 1 as necessary.

[0018] Typically, a situation where remote assistance by an operator is necessary is a situation where autonomous driving is difficult. The autonomous driving vehicle 1 executes various vehicle processes during autonomous driving. Representative vehicle processes during autonomous driving include the following: (1) Recognition process: The autonomous vehicle 1 uses a recognition sensor to recognize the situation around the autonomous vehicle 1. For example, the autonomous vehicle 1 uses a camera to recognize the signal display of a traffic light (e.g., green light, yellow light, red light, right turn signal, etc.). (2) Action Determination Process: Based on the result of the recognition process, the autonomous vehicle 1 determines whether or not to execute an action. Examples of the actions of the autonomous vehicle 1 include starting, stopping, turning right, turning left, changing lanes, etc. (3) Timing determination process: The autonomously driven vehicle 1 determines the execution timing for executing the above action.

[0019] For example, when sunlight is shining on a traffic light installed at an intersection, the accuracy of recognizing the traffic light display may decrease. If the traffic light display cannot be accurately determined by the recognition process, the autonomous vehicle 1 requires remote assistance from an operator for traffic light recognition. In addition, if the traffic light display cannot be determined, it is difficult to determine what action should be performed and when. Therefore, the autonomous vehicle 1 requires remote assistance from an operator for the action determination process and the timing determination process as well.

[0020] Even if the signal display is identified, there may be a situation where it is difficult to determine whether or not it is acceptable to actually execute an action. For example, even after the signal display as seen by the autonomous vehicle 1 indicates "right turn permitted," an oncoming vehicle may enter the intersection, or an oncoming vehicle or a preceding vehicle may remain in the intersection. In such a case, the autonomous vehicle 1 may request remote assistance from an operator for the action determination process and timing determination process while remaining stopped.

[0021] As yet another example, there may be a situation where a parked vehicle is present in the lane ahead of the autonomous vehicle 1, but it is difficult to determine whether or not to change lanes. For example, in a situation where there are signs that the parked vehicle is about to depart, it is difficult to determine whether or not to change lanes. Also, in a section where lane changes are prohibited, there is no action that can be taken other than stopping. In such a case, the autonomous vehicle 1 may stop in front of the parked vehicle and request remote assistance from an operator for the action determination process.

[0022] As yet another example, in a construction zone, the autonomous vehicle 1 may request remote assistance from an operator. As yet another example, in the event of a breakdown in the autonomous vehicle 1, the autonomous vehicle 1 may request remote assistance from an operator.

[0023] When remote assistance from an operator is required, the autonomous vehicle 1 sends an "assistance request REQ" to the remote assistance system 100. The assistance request REQ requests assistance with at least one of the multiple vehicle processes described above (recognition process, action determination process, and timing determination process). This process by the autonomous vehicle 1 is hereinafter referred to as "assistance request process." In addition, the autonomous vehicle 1 that performs the assistance request process is hereinafter referred to as the "target autonomous vehicle 1T." The assistance request REQ may include the results of the vehicle processing in the target autonomous vehicle 1T as reference information.

[0024] The remote assistance system 100 receives an assistance request REQ from the target autonomously driven vehicle 1T. The remote assistance system 100 notifies the received assistance request REQ to an operator and requests the operator to provide instructions for the target autonomously driven vehicle 1T. This processing by the remote assistance system 100 will be referred to hereinafter as "operator instruction request processing."

[0025] The operator determines an "operator instruction INS", which is an instruction for the target autonomously driven vehicle 1T, according to the contents of the assistance request REQ. The operator inputs the operator instruction to the remote assistance system 100. The remote assistance system 100 communicates with the target autonomously driven vehicle 1T and notifies the target autonomously driven vehicle 1T of the operator instruction INS. This processing by the remote assistance system 100 is hereinafter referred to as "operator instruction notification processing".

[0026] The remote support processing by the remote support system 100 includes the above-mentioned "operator instruction request processing" and "operator instruction notification processing."

[0027] The target autonomously driven vehicle 1T receives the operator instruction INS from the remote assistance system 100. Then, the target autonomously driven vehicle 1T resumes autonomous driving in accordance with the operator instruction INS.

[0028] Incidentally, there may be cases where autonomous driving is still difficult even after receiving the operator instruction INS. In this case, the target autonomous driving vehicle 1T may resend the assistance request REQ to the remote assistance system 100 without following the received operator instruction INS. This process will be referred to as "re-assistance request process" below.

[0029] 2. Reducing the burden on operators 2-1. Overview As described above, the operator remotely supports the driving of the target autonomous vehicle 1T in response to an assistance request REQ from the target autonomous vehicle 1T. At this time, if the burden on the operator increases, the efficiency of the remote assistance may decrease.

[0030] For example, consider a situation in which assistance requests REQ are sent one after another from a large number of target autonomous vehicles 1T. The operator needs to quickly handle the assistance requests REQ that occur one after another. If the response to the assistance requests REQ is delayed, the efficiency of remote assistance decreases.

[0031] It is possible to prepare a large number of operators and have them divide the work among them to handle the support requests REQ. However, the operators who provide remote support are highly skilled personnel who require specialized skills. There are limitations in terms of cost and education in terms of increasing the number of such highly skilled operators.

[0032] In order to improve the efficiency of remote assistance, it is important to reduce the burden on the operator. To that end, it is effective to reduce the number of matters that the operator must consider and decide for a single assistance request REQ. This reduces the time required for the operator to deal with a single assistance request REQ (i.e., a single target autonomous vehicle 1T). As a result, the efficiency of remote assistance is improved.

[0033] From the above perspective, according to this embodiment, the concepts of "assistance processing" and "delegated processing" are introduced. An assistance processing is a vehicle processing among the above-mentioned multiple vehicle processing (recognition processing, behavior determination processing, and timing judgment processing) that is performed by an operator on behalf of the target autonomous vehicle 1T. On the other hand, a delegated processing is a processing among the above-mentioned multiple vehicle processing (recognition processing, behavior determination processing, and timing judgment processing) that the operator leaves to the target autonomous vehicle 1T.

[0034] As the number of delegated processes increases and the number of support processes decreases, the number of items that the operator must decide decreases accordingly. This is preferable from the viewpoint of reducing the burden on the operator. However, in order to determine which vehicle processes should be delegated processes, various pieces of information must be comprehensively considered. Moreover, the various pieces of information to be considered differ for each target autonomous vehicle 1T and also differ depending on the situation in which the target autonomous vehicle 1T is placed. This consideration ultimately places a burden on the operator.

[0035] Therefore, the remote assistance system 100 according to the present embodiment presents to the operator useful information that contributes to shortening the consideration time of the operator in the operator instruction request process. There are multiple options OPT for combinations of assistance processes and delegation processes. The remote assistance system 100 automatically determines one of the multiple options OPT to be recommended to the operator as a "recommended option OPT1". Then, the remote assistance system 100 presents at least the recommended option OPT1 to the operator. By referring to the recommended option OPT1, the operator can shorten the time it takes to consider which vehicle process to delegation process. This reduces the burden on the operator.

[0036] 2-2. Remote support processing based on performance database 3 is a conceptual diagram for explaining remote assistance processing by the remote assistance system 100 according to this embodiment. The remote assistance system 100 includes a track record database 200. The track record database 200 shows at least past track records of assistance request processing by each autonomously driving vehicle 1 for each scene.

[0037] A scene is a situation in which the autonomous vehicle 1 is placed. Specifically, a scene includes locations where an assistance request process may be performed by the autonomous vehicle 1. Examples of locations where an assistance request process may be performed include intersections, merging sections, construction sections, etc. Each location is specified by a range of latitude and longitude. A scene may include not only a location but also an action of the autonomous vehicle 1. Examples of a scene including a location and an action include "making a right turn at an intersection" and "changing lanes at a merging section."

[0038] The track record of assistance request processing in a certain scene includes, for example, the probability that assistance from vehicle processing was requested in that scene. For example, assume that a certain autonomous vehicle 1 has passed a certain intersection 100 times in the past. Assume that the recognition processing to recognize a traffic light display at that intersection has been successful 98 times and failed twice. If recognition of the traffic light display fails, the autonomous vehicle 1 requests assistance from the recognition processing. Therefore, the past assistance request probability for the recognition processing at that intersection is 2%. If the past assistance request probability for the recognition processing is low, it is considered that problems are unlikely to occur even if the recognition processing is assigned to a delegated processing.

[0039] In this way, the track record database 200, which shows the past track records of assistance request processing for each scene, is useful for determining the recommended option OPT1. In the above example, the assistance request probability is 2%, and the recognition success probability is 98%. It can be said that the "track record of assistance request processing" and the "track record of vehicle processing" are in an inverse relationship and are equivalent. Therefore, it can also be said that the track record database 200 shows the track records of multiple vehicle processing by the autonomous vehicle 1 for each scene.

[0040] Each time each autonomous vehicle 1 experiences a scene, it sends information indicating the scene and the results of each vehicle process (recognition process, action determination process, and timing determination process) to the remote assistance system 100. When an assistance request process is performed, each autonomous vehicle 1 sends information indicating the contents of the assistance request process (which vehicle process assistance was requested) to the remote assistance system 100. The remote assistance system 100 collects this information from each autonomous vehicle 1. The remote assistance system 100 then updates the track record database 200 by registering the collected information in the track record database 200. As shown in FIG. 3, the track record database 200 indicates the correspondence between the identification information of each autonomous vehicle 1, the scene, and the track record of the assistance request process.

[0041] Meanwhile, an assistance request REQ is sent from a target autonomous vehicle 1T in a certain scene to the remote assistance system 100. For convenience, the scene in which the target autonomous vehicle 1T is located is referred to as a "first scene." In response to the assistance request REQ, the remote assistance system 100 performs an operator instruction request process that requests the operator to give instructions to the target autonomous vehicle 1T.

[0042] In the operator instruction request processing, the remote assistance system 100 first obtains the results of the assistance request processing related to the first scene from the results database 200. The assistance request processing related to the first scene includes the assistance request processing in the first scene itself. The assistance request processing related to the first scene may include the assistance request processing in a similar scene similar to the first scene. For example, if the first scene is "making a right turn at intersection A," then "making a right turn at intersection B, which has the same structure as intersection A," is a similar scene.

[0043] The track record of the assistance request processing acquired from the track record database 200 includes at least the track record of the assistance request processing performed by the target autonomously driven vehicle 1T itself. In other words, the remote assistance system 100 acquires from the track record database 200 the track record of the assistance request processing related to the first scene performed by the target autonomously driven vehicle 1T.

[0044] As a variant example, the remote assistance system 100 may acquire the track record of assistance request processing by another autonomously driven vehicle 1, in addition to the track record of assistance request processing by the target autonomously driven vehicle 1T. In this case, the remote assistance system 100 combines the track records of assistance request processing by the target autonomously driven vehicle 1T and the other autonomously driven vehicle 1 after weighting them. At this time, the weight for the target autonomously driven vehicle 1T is set to be larger than the weight for the other autonomously driven vehicle 1.

[0045] Next, the remote assistance system 100 determines a "recommended option OPT1" for a combination of an assistance process and a delegated process based on the track record of the assistance request process related to the first scene. The assistance process is a vehicle process that the operator performs on behalf of the target autonomously driven vehicle 1T among the multiple vehicle processes in the first scene. On the other hand, the delegated process is a process that the operator leaves to the target autonomously driven vehicle 1T among the multiple vehicle processes in the first scene. Then, the remote assistance system 100 presents at least the recommended option OPT1 to the operator. The remote assistance system 100 may present a plurality of options OPT including the recommended option OPT1 to the operator.

[0046] Fig. 4 is a conceptual diagram for explaining an example of an operator instruction request process according to this embodiment. Specifically, Fig. 4 shows an example of information displayed on a monitor viewed by an operator. An image (video) captured by a camera mounted on a target autonomous driving vehicle 1T is displayed on the monitor. This image information is transmitted in real time from the target autonomous driving vehicle 1T to the remote assistance system 100 via communication. The remote assistance system 100 displays the image information on the monitor.

[0047] In the example shown in FIG. 4, the target autonomous vehicle 1T issues an assistance request REQ when making a right turn at the first intersection. That is, the first scene is a right turn at the first intersection. The target autonomous vehicle 1T performs recognition processing and recognizes the signal display of a traffic light installed at the first intersection. In this example, the target autonomous vehicle 1T recognizes the signal display as a "right arrow" with a high degree of certainty. Therefore, the target autonomous vehicle 1T determines that remote assistance from an operator is not required for signal recognition.

[0048] On the other hand, the target autonomous vehicle 1T cannot confidently determine whether to make a right turn. As a result, it cannot determine the timing of making a right turn. In such a situation, the target autonomous vehicle 1T sends an assistance request REQ to the remote assistance system 100. The assistance request REQ requests assistance with the action determination process (right turn determination process) and the timing determination process. The assistance request REQ may include the results of the recognition process (signal recognition), the action determination process, and the timing determination process in the target autonomous vehicle 1T. In that case, each result is also displayed on the monitor as shown in FIG. 4.

[0049] In the operator instruction request processing, the remote assistance system 100 acquires the results of the assistance request processing related to the first scene from the results database 200. For example, the results of each vehicle processing and assistance request processing by the target autonomous driving vehicle 1T in the first scene are as follows. [Recognition processing (signal recognition)] Success rate: 98%, assistance request rate: 2% [Action decision process (right turn decision)] Success probability = 40%, assistance request probability = 60% [Timing judgment process] Success probability = 95%, Support request probability = 5%

[0050] For vehicle processing with a sufficiently low past probability of requesting assistance, it is considered acceptable to delegate the processing to the target autonomous vehicle 1T. Therefore, the remote assistance system 100 recommends to the operator that the timing judgment processing be delegated to the target autonomous vehicle 1T. On the other hand, for vehicle processing with a high past probability of requesting assistance, it is desirable for the operator to provide assistance. Therefore, the remote assistance system 100 recommends to the operator that the operator assist with the behavior judgment processing. In other words, the recommended option OPT1 is "assist with the behavior judgment processing, but leave the timing judgment processing to the target autonomous vehicle 1T."

[0051] The remote assistance system 100 presents at least the recommended option OPT1 to the operator. For example, as shown in FIG. 4, the recommended option OPT1 is displayed on the operator instruction screen. The recommended option OPT1 may be accompanied by the word "recommended." By referring to the recommended option OPT1, the operator can reduce the time it takes to consider which vehicle process to delegate. This reduces the burden on the operator.

[0052] The operator does not necessarily have to follow the recommended option OPT1. In some cases, the operator may give instructions different from the recommended option OPT1. Even in this case, the effect of shortening the examination time and reducing the load can still be obtained.

[0053] As shown in FIG. 4, multiple options OPT including the recommended option OPT1 may be presented to the operator. For example, the option OPT2 is "support both the action determination process and the timing determination process." The option OPT3 is "support all of the recognition process, the action determination process, and the timing determination process." The option OPT3 suggests that the operator also assist with the recognition process for which assistance has not been requested by the target autonomous driving vehicle 1T. When multiple options OPT including the recommended option OPT1 are presented to the operator, the operator can easily compare the multiple options OPT. This also contributes to shortening the consideration time.

[0054] When multiple options OPT are presented to the operator, it is preferable that the recommended option OPT1 is presented in a more prominent manner than the other options. For example, as shown in Fig. 4, the character size of the recommended option OPT1 is larger than the character sizes of the other options OPT2 and OPT3. The characters of the recommended option OPT1 may be bolder than the characters of the other options OPT2 and OPT3. This makes it easier for the operator to understand the recommended option OPT1.

[0055] 2-3.Effects As described above, according to this embodiment, the remote assistance system 100 requests the operator to give instructions to the target autonomously driven vehicle 1T in response to an assistance request REQ from the target autonomously driven vehicle 1T. The assistance processing is vehicle processing performed by the operator on behalf of the target autonomously driven vehicle 1T. On the other hand, the delegated processing is processing that the operator leaves to the target autonomously driven vehicle 1T. As the delegated processing increases and the assistance processing decreases, the number of items that the operator must decide decreases accordingly. This contributes to reducing the burden on the operator.

[0056] Furthermore, according to this embodiment, the remote assistance system 100 determines a recommended option OPT1 for a combination of assistance processing and delegated processing based on the result database 200 indicating the results of assistance request processing. Then, the remote assistance system 100 presents the recommended option OPT1 to the operator. By referring to the recommended option OPT1, the operator can reduce the time it takes to consider which vehicle processing to delegate. This reduces the burden on the operator.

[0057] Since the burden on the operator is reduced, the time required for the operator to deal with a single assistance request REQ (i.e., a single target autonomously driven vehicle 1T) is shortened. As a result, the efficiency of remote assistance is improved. This also means that a limited number of operators can handle a larger number of assistance requests REQ (target autonomously driven vehicles 1T). This is preferable from the viewpoint of cost. According to this embodiment, it is possible to improve the efficiency of remote assistance while suppressing an increase in costs.

[0058] Note that the remote assistance system 100 according to this embodiment does not perform remote assistance completely automatically without the intervention of an operator. The final instruction content (operator instruction INS) for the target autonomous driving vehicle 1T is determined by the operator. This ensures the accuracy of the remote assistance. In addition, the operator does not necessarily have to follow the recommended option OPT1. In some cases, the operator may give an instruction different from the recommended option OPT1. This ensures the flexibility of the remote assistance.

[0059] A specific example of the autonomously driving vehicle 1 and the remote assistance system 100 according to this embodiment will be described in detail below.

[0060] 3. Self-driving vehicles 3-1. Configuration example 5 is a block diagram showing an example of the configuration of an autonomous driving system 10 mounted on an autonomous driving vehicle 1 according to this embodiment. The autonomous driving system 10 includes a recognition sensor 20, a vehicle state sensor 30, a position sensor 40, a communication device 50, a driving device 60, and a control device 70.

[0061] The recognition sensor 20 recognizes (detects) the surrounding situation of the autonomous driving vehicle 1. The recognition sensor 20 includes at least a camera 25. The recognition sensor 20 may include a LIDAR (Laser Imaging Detection and Ranging), a radar, or the like.

[0062] The vehicle state sensor 30 detects the state of the autonomously driven vehicle 1. For example, the vehicle state sensor 30 includes a vehicle speed sensor, an acceleration sensor, a yaw rate sensor, a steering angle sensor, and the like.

[0063] The position sensor 40 detects the position and orientation of the autonomously driven vehicle 1. An example of the position sensor 40 is a Global Positioning System (GPS) sensor.

[0064] The communication device 50 communicates with the outside of the autonomous driving vehicle 1. For example, the communication device 50 communicates with the remote assistance system 100.

[0065] The traveling device 60 includes a steering device, a drive device, and a braking device. The steering device steers the wheels of the autonomously driven vehicle 1. For example, the steering device includes an electric power steering (EPS) device. The drive device is a power source that generates a driving force. Examples of the drive device include an engine, an electric motor, and an in-wheel motor. The braking device generates a braking force.

[0066] The control device 70 controls the autonomous vehicle 1. The control device 70 includes one or more processors 71 (hereinafter simply referred to as the processor 71) and one or more memories 72 (hereinafter simply referred to as the memory 72). The processor 71 executes various processes. For example, the processor 71 includes a CPU (Central Processing Unit). The memory 72 stores various information. Examples of the memory 72 include a volatile memory, a non-volatile memory, a HDD (Hard Disk Drive), and an SSD (Solid State Drive). The processor 71 executes a control program, which is a computer program, to thereby realize various processes by the processor 71 (control device 70). The control program is stored in the memory 72 or is recorded in a computer-readable recording medium. The control device 70 may include one or more ECUs (Electronic Control Units). A part of the control device 70 may be an information processing device external to the autonomous vehicle 1. In that case, a part of the control device 70 communicates with the autonomous vehicle 1 and remotely controls the autonomous vehicle 1.

[0067] 3-2. Information Acquisition Processing The processor 71 acquires driving environment information 80 that indicates the driving environment of the autonomously driven vehicle 1. The driving environment information 80 is stored in the memory 72.

[0068] 6 is a block diagram showing an example of the driving environment information 80. The driving environment information 80 includes map information 81, surrounding situation information 82, vehicle state information 83, and vehicle position information 84.

[0069] Map information 81 indicates lane arrangement, road shape, etc. Processor 71 acquires map information of a required area from a map database. The map database may be stored in a predetermined storage device mounted on autonomous vehicle 1, or may be stored in a management server external to autonomous vehicle 1. In the latter case, processor 71 communicates with the management server to acquire the required map information.

[0070] The surrounding situation information 82 is information that indicates the situation around the autonomous driving vehicle 1. The surrounding situation information 82 includes information obtained by the recognition sensor 20. For example, the surrounding situation information 82 includes image information IMG captured by the camera 25. As another example, the surrounding situation information 82 includes measurement information measured by a LIDAR or a radar.

[0071] Moreover, the surrounding situation information 82 includes object information on objects around the autonomous vehicle 1. Examples of objects around the autonomous vehicle 1 include pedestrians, other vehicles (vehicles ahead, parked vehicles, etc.), signs, white lines, roadside structures, obstacles, etc. The object information indicates the relative position and relative speed of the object with respect to the autonomous vehicle 1. For example, by analyzing the image information IMG obtained by the camera 25, it is possible to identify the object and calculate the relative position of the object. It is also possible to identify the object and calculate the relative position and relative speed of the object based on measurement information by LIDAR or radar.

[0072] Furthermore, the surrounding situation information 82 includes traffic light display information indicating the recognition result of the traffic light signal display. The processor 71 detects traffic lights around the autonomously driven vehicle 1 based on the image information IMG captured by the camera 25. Furthermore, the processor 71 recognizes the traffic light signal display (e.g., green light, yellow light, red light, right turn signal, etc.) based on the image information IMG. Image analysis methods for detecting (extracting) traffic lights from an image and recognizing the signal display are well known.

[0073] The vehicle state information 83 is information indicating the state of the autonomously driven vehicle 1. Examples of the state of the autonomously driven vehicle 1 include the vehicle speed, yaw rate, lateral acceleration, steering angle, etc. The processor 71 acquires the vehicle state information 83 from the detection result by the vehicle state sensor 30.

[0074] The vehicle position information 84 indicates the position and orientation of the autonomously driving vehicle 1. The vehicle position information 84 is obtained by the position sensor 40. The processor 71 may also obtain highly accurate vehicle position information 84 by a well-known self-position estimation process (Localization) using map information 81 and surrounding situation information 82 (object information).

[0075] 3-3.Vehicle driving control The processor 71 executes "vehicle driving control" that controls the driving of the autonomously driven vehicle 1. The vehicle driving control includes steering control, acceleration control, and deceleration control. The processor 71 executes the vehicle driving control by controlling the driving device 60. Specifically, the processor 71 executes the steering control by controlling the steering device. The processor 71 also executes the acceleration control by controlling the drive device. The processor 71 also executes the deceleration control by controlling the braking device.

[0076] 3-4.Automatic driving control The processor 71 performs autonomous driving control based on the driving environment information 80. The autonomous driving here is assumed to be based on the premise that the driver does not necessarily have to concentrate 100% on driving (so-called autonomous driving of level 3 or higher).

[0077] During the autonomous driving, the processor 71 executes the vehicle processing described above. The vehicle processing includes the following: (1) Recognition process: The processor 71 uses the recognition sensor 20 to recognize the surrounding situation of the autonomous vehicle 1. For example, the processor 71 uses the camera 25 to recognize the signal display of a traffic light (e.g., green light, yellow light, red light, right turn signal, etc.). (2) Action Determination Process: The processor 71 determines whether or not to execute an action based on the result of the recognition process. Examples of the actions of the autonomous driving vehicle 1 include starting, stopping, turning right, turning left, changing lanes, and the like. (3) Timing determination process: Processor 71 determines the execution timing for executing the above action.

[0078] When performing the above action, the processor 71 generates a target trajectory for the autonomous vehicle 1 based on the driving environment information 80. The target trajectory includes a target position and a target speed. Then, the processor 71 executes vehicle driving control so that the autonomous vehicle 1 follows the target trajectory.

[0079] 3-5. Support request processing When remote assistance from an operator is required, the processor 71 sends an assistance request REQ to the remote assistance system 100. More specifically, the processor 71 communicates with the remote assistance system 100 via the communication device 50 and sends the assistance request REQ to the remote assistance system 100. Typically, a situation in which remote assistance from an operator is required is a situation in which autonomous driving is difficult. Therefore, the assistance request REQ requests assistance in at least one of the above-mentioned vehicle processes (recognition process, action judgment process, and timing judgment process). The assistance request REQ may include the result of the vehicle process as reference information.

[0080] The processor 71 sends driving environment information 80 necessary for remote assistance to the remote assistance system 100 together with the assistance request REQ. The driving environment information 80 necessary for remote assistance includes at least image information IMG captured by the camera 25. The processor 71 sends the latest image information IMG to the remote assistance system 100. The processor 71 may also send image information IMG for a certain period before the assistance request REQ is generated to the remote assistance system 100. In addition, the processor 71 may send necessary vehicle state information 83, etc. to the remote assistance system 100.

[0081] 3-6. Automatic operation control according to operator instructions After sending the assistance request REQ, the processor 71 receives an operator instruction INS from the remote assistance system 100 via the communication device 50. The operator instruction INS includes instructions from the operator regarding the assistance process. The operator instruction INS may also include a designation of a delegated process. The processor 71 resumes the automatic driving control in accordance with the operator instruction INS. At this time, with regard to the vehicle process designated as the delegated process, the processor 71 recognizes or judges it by itself without relying on instructions from the operator.

[0082] 3-7. Processing of re-support requests There may be cases where automatic driving is still difficult even after receiving the operator instruction INS. In that case, the processor 71 does not follow the received operator instruction INS and sends the assistance request REQ again to the remote assistance system 100. The assistance request REQ may indicate that "this is a second assistance request."

[0083] 4. Remote Support System 4-1. Configuration example 7 is a block diagram showing an example of the configuration of a remote support system 100 according to this embodiment. The remote support system 100 includes an information processing device 110, a communication device 120, a display device 130, an input device 140, and a performance database 200.

[0084] The information processing device 110 performs various information processing. The information processing device 110 includes one or more processors 111 (hereinafter simply referred to as processor 111) and one or more memories 112 (hereinafter simply referred to as memory 112). The processor 111 executes various processes. For example, the processor 111 includes a CPU. The memory 112 stores various information. Examples of the memory 112 include a volatile memory, a non-volatile memory, a HDD, and an SSD. The processor 111 executes a remote support program, which is a computer program, to realize the functions of the information processing device 110. The remote support program is stored in the memory 112. The remote support program may be recorded in a computer-readable recording medium. The remote support program may be provided via a network.

[0085] The communication device 120 communicates with the outside. For example, the communication device 120 communicates with the autonomously driven vehicle 1.

[0086] The display device 130 displays various information. Examples of the display device 130 include a liquid crystal display, an organic EL display, a head mounted display, a touch panel, etc. The display device 130 corresponds to the monitor shown in FIG.

[0087] The input device 140 is an interface for receiving input from an operator. Examples of the input device 140 include a touch panel, a keyboard, and a mouse.

[0088] 4-2. Performance database The information processing device 110 can access a performance database 200. The performance database 200 indicates the performance of vehicle processing and assistance request processing by each autonomously driving vehicle 1 for each scene.

[0089] 8 is a conceptual diagram showing an example of the result database 200. The result database 200 shows the correspondence between vehicle ID information, scenes, and results.

[0090] The vehicle ID information includes an ID number assigned to each autonomously driven vehicle 1. The vehicle ID information may also include the vehicle type (e.g., standard, medium, large, bus, etc.) of each autonomously driven vehicle 1.

[0091] A scene includes locations where an assistance request process may be performed by the autonomous vehicle 1. Locations where an assistance request process may be performed include intersections, merging sections, construction sections, etc. Each location is defined by a range of latitude and longitude. A scene may include not only locations but also actions of the autonomous vehicle 1.

[0092] The track record includes, for example, a support request probability for each of a plurality of vehicle processes. The support request probability is a ratio of the number of occurrences of the support request process to the number of times the autonomous vehicle 1 has experienced the scene. Note that the support request probability and the success probability of the vehicle process are inversely related and are equivalent. In the example shown in FIG. 8, the track record indicates the success probability of each of a plurality of vehicle processes.

[0093] As another example, the track record may include a frequency of assistance requests for each of a plurality of vehicle processes, the frequency of assistance requests being calculated based on the number of occurrences of assistance requests in a certain period of time in the past.

[0094] Fig. 9 is a conceptual diagram showing another example of the result database 200. In the example shown in Fig. 9, the results shown in the result database 200 further include the results of re-support request processing. For example, the result of re-support request processing is a re-support request probability. The re-support request probability is a ratio of the occurrence frequency of re-support request processing to the frequency of operator instruction INS.

[0095] As another example, the performance of the re-support request process may include a frequency of re-support requests, which is calculated based on the number of occurrences of the re-support request process in a certain period in the past.

[0096] Each time an autonomous vehicle 1 experiences a certain scene, it sends information indicating the scene and the results of each vehicle process to the remote assistance system 100. When an assistance request process is performed, each autonomous vehicle 1 sends information indicating the contents of the assistance request process (which vehicle process assistance was requested) to the remote assistance system 100. The processor 111 receives information from the autonomous vehicle 1 via the communication device 120. The processor 111 updates the performance database 200 by registering the collected information in the performance database 200.

[0097] 4-3. Remote support processing The processor 111 executes remote assistance processing. The remote assistance processing includes an "operator instruction request processing" that requests an operator to give instructions to the target autonomously driven vehicle 1T in response to an assistance request REQ sent from the target autonomously driven vehicle 1T. The remote assistance processing also includes an "operator instruction notification processing" that notifies the target autonomously driven vehicle 1T of an operator instruction INS from the operator.

[0098] 10 is a block diagram showing various types of information related to the remote assistance process according to the present embodiment. The various types of information are stored in the memory 112.

[0099] The assistance request REQ is sent from the target autonomously driven vehicle 1T. The processor 111 receives the assistance request REQ via the communication device 120.

[0100] The vehicle side information 300 is driving environment information 80 necessary for remote assistance. The vehicle side information 300 includes at least image information IMG captured by the camera 25. The vehicle side information 300 may include not only the latest image information IMG but also image information IMG for a certain period before the assistance request REQ is generated. The vehicle side information 300 may include vehicle state information 83. The vehicle side information 300 is sent from the target autonomous driving vehicle 1T. The processor 111 receives the vehicle side information 300 via the communication device 120.

[0101] The reference information 400 is information that is referred to when determining the recommended option OPT1.

[0102] The reference information 400 includes performance information 410. The performance information 410 is information obtained from the performance database 200. Specifically, the performance information 410 indicates performance related to a first scene in which the target autonomously driven vehicle 1T is placed. The processor 111 accesses the performance database 200 and obtains the performance information 410 from the performance database 200.

[0103] The reference information 400 may include the autonomous driving system information 430. The autonomous driving system information 430 indicates the state of the autonomous driving system 10. For example, the autonomous driving system information 430 indicates whether the autonomous driving software has just been updated. The autonomous driving system information 430 may indicate the accuracy of a sensor such as the recognition sensor 20. The autonomous driving system information 430 may indicate whether the traveling device 60 is normal or not. The autonomous driving system information 430 is sent from the target autonomous driving vehicle 1T. The processor 111 receives the autonomous driving system information 430 via the communication device 120.

[0104] The reference information 400 may include environmental condition information 440. The environmental condition information 440 indicates the environment in which the target autonomous vehicle 1T is located. Examples of the environment include weather, meteorological conditions, time, season, traffic congestion, and the like. The environmental condition information 440 is provided from an information service server or the target autonomous vehicle 1T. The processor 111 receives the environmental condition information 440 via the communication device 120.

[0105] The option information 500 indicates a plurality of options OPT for combinations of support processing and delegation processing. The plurality of options OPT includes a recommended option OPT1. The processor 111 generates the option information 500 by processing to be described later.

[0106] The operator instruction INS is an instruction from the operator to the target autonomously driven vehicle 1T. The operator instruction INS includes an instruction from the operator regarding the assistance process. The operator instruction INS may also include a designation of a delegated process. The operator inputs the operator instruction INS using the input device 140. The processor 111 receives the operator instruction INS via the input device 140.

[0107] FIG. 11 is a flowchart showing the operator instruction request process in the remote support process.

[0108] In step S110, the processor 111 determines whether or not an assistance request REQ has been received from the target autonomously driven vehicle 1T. If an assistance request REQ has been received (step S110; Yes), the process proceeds to step S120. Otherwise (step S110; No), the process in the current cycle ends.

[0109] In step S120 , the processor 111 acquires the vehicle side information 300 and the reference information 400 .

[0110] In step S130, the processor 111 displays the vehicle-side information 300 on the display device 130. In particular, the processor 111 displays the image information IMG on the display device 130 (see FIG. 4).

[0111] In step S140, the processor 111 sets a plurality of options OPT for combinations of support processing and delegation processing. Furthermore, the processor 111 determines a recommended option OPT1 to be recommended to the operator from the plurality of options OPT. Here, the processor 111 determines the recommended option OPT1 based on the above-mentioned reference information 400. Various examples of determining the recommended option OPT1 based on the reference information 400 will be described later. The processor 111 generates option information 500 through this processing.

[0112] In step S150, the processor 111 displays the option information 500 on the display device 130. At this time, the processor 111 displays at least the recommended option OPT1. The processor 111 may display a plurality of options OPT including the recommended option OPT1. The processor 111 may display the plurality of options OPT so that the recommended option OPT1 stands out more than the other options OPT (see FIG. 4).

[0113] FIG. 12 is a flowchart showing an operator instruction notification process in the remote assistance process.

[0114] In step S160, processor 111 determines whether or not an operator instruction INS has been input by the operator. If an operator instruction INS has been input (step S160; Yes), the process proceeds to step S170. Otherwise (step S160; No), the process in the current cycle ends.

[0115] In step S170, the processor 111 communicates with the target autonomously driven vehicle 1T via the communication device 120 and notifies the target autonomously driven vehicle 1T of the operator instructions INS.

[0116] 4-4. Example of recommended option determination process Various examples of the recommended option determination process (step S140) will be described below.

[0117] 4-4-1. First example In a first example, the processor 111 determines a recommended option OPT1 based on performance information 410. The performance information 410 indicates a performance associated with a first scene in which a target autonomous vehicle 1T is located.

[0118] For example, the track record includes an assistance request probability or an assistance request frequency for each of the multiple vehicle processes. Any one of the multiple vehicle processes is referred to as a "first vehicle process" for convenience. Processor 111 acquires the assistance request probability or the assistance request frequency for the first vehicle process from track record information 410. Then, based on the assistance request probability or the assistance request frequency for the first vehicle process, processor 111 determines whether to set the first vehicle process as an assistance process or a delegated process in the recommended option OPT1.

[0119] For example, as the assistance request probability or assistance request frequency for the first vehicle process decreases, the probability that the first vehicle process is set as a delegated process in the recommended option OPT1 increases. Conversely, as the assistance request probability or assistance request frequency for the first vehicle process increases, the probability that the first vehicle process is set as a assisted process in the recommended option OPT1 increases.

[0120] As another example, the assistance request probability or the assistance request frequency may be compared to a predetermined threshold. If the assistance request probability or the assistance request frequency for the first vehicle process is equal to or greater than the predetermined threshold, the processor 111 sets the first vehicle process as an assistance process in the recommended option OPT1. On the other hand, if the assistance request probability or the assistance request frequency for the first vehicle process is less than the predetermined threshold, the processor 111 sets the first vehicle process as a delegated process in the recommended option OPT1.

[0121] In this manner, the contents of the recommended option OPT1 are determined based on the performance information 410. Since the performance information 410 indicates past performance regarding support request processing, it becomes possible to determine an appropriate recommended option OPT1.

[0122] 4-4-2. Second example In the second example, the performance of the re-support request processing is taken into consideration. The processor 111 acquires performance information 410 (see FIG. 9) including the performance of the re-support request processing, and determines the recommended option OPT1 based on the performance of the re-support request processing.

[0123] For example, the performance of the re-support request processing includes the re-support request probability or the re-support request frequency. The processor 111 acquires the re-support request probability or the re-support request frequency from the performance information 410. Then, the processor 111 sets the support processing and the delegation processing in the recommended option OPT1 based on the re-support request probability or the re-support request frequency. For example, as the re-support request probability or the re-support request frequency increases, the ratio of the support processing in the recommended option OPT1 increases and the ratio of the delegation processing decreases. Conversely, as the re-support request probability or the re-support request frequency decreases, the ratio of the delegation processing in the recommended option OPT1 increases and the ratio of the support processing decreases.

[0124] 4-4-3.Third example In a third example, the processor 111 determines a recommended option OPT1 based on the automated driving system information 430.

[0125] For example, immediately after updating the autonomous driving software, the reliability of past performance is reduced. Therefore, in this case, the processor 111 increases the ratio of assistance processing in the recommendation option OPT1 and decreases the ratio of delegation processing.

[0126] As another example, when the accuracy of a sensor such as the recognition sensor 20 is degraded, the accuracy of the automatic driving control is also degraded. Therefore, in this case, the processor 111 increases the ratio of the support process in the recommended option OPT1 and decreases the ratio of the delegation process.

[0127] 4-4-4. Fourth Example In a fourth example, the processor 111 determines the recommended option OPT1 based on the environmental condition information 440.

[0128] For example, in bad weather conditions, the accuracy of the recognition process may decrease, and the accuracy of the automatic driving control may also decrease. Therefore, the processor 111 increases the ratio of the support process in the recommended option OPT1 and decreases the ratio of the delegation process.

[0129] 4-4-5. Fifth Example It is also possible to combine two or more of the first to fourth examples. [Explanation of symbols]

[0130] 1. Self-driving vehicles 1T Target autonomous vehicle 10 Autonomous Driving Systems 20 Recognition Sensor 25 Camera 50 Communication equipment 70 Control device 80 Driving Environment Information 100 Remote Support System 110 Information processing device 111 Processor 112 Memory 120 Communication Equipment 130 Display device 140 Input Devices 200 Achievement database 300 Vehicle information 400 References 410 Performance Information 430 Autonomous Driving System Information 440 Environmental condition information 500 Choice Information IMG Image Information INS Operator Instructions OPT1 Recommended option REQ Request for assistance

Claims

1. A remote assistance system that remotely assists the driving of an autonomous vehicle, The autonomous vehicle includes: A plurality of vehicle processes including recognition and decision making during autonomous driving; an assistance request process for sending an assistance request to the remote assistance system, the assistance request requesting assistance for at least one of the plurality of vehicle processes; Run The remote assistance system includes: one or more processors; a performance database showing the performance of the assistance request processing by the autonomous driving vehicle for each scene; Equipped with The one or more processors: an operator instruction request process that requests an operator to provide instructions for a target autonomous vehicle in a first scene in response to the assistance request sent from the target autonomous vehicle in a first scene; an operator instruction notification process for notifying the target autonomous driving vehicle of the instruction from the operator; configured to run The assistance process is a vehicle process among the plurality of vehicle processes in the first scene that is performed by the operator instead of the target autonomous vehicle, A delegated process is a vehicle process among the plurality of vehicle processes in the first scene that the operator entrusts to the target autonomous vehicle, The target autonomous vehicle executes the delegation process without relying on instructions from the operator; and In the operator instruction request processing, the one or more processors: obtaining the performance of the assistance request processing related to the first scene from the performance database; determining a recommended option to be recommended to the operator from among a plurality of options of combinations of the support process and the delegation process based on the track record related to the first scene; Presenting at least the recommended options to the operator Remote support system.

2. The remote assistance system according to claim 1, The one or more processors present the plurality of options, including the recommended option, to the operator. Remote support system.

3. The remote assistance system according to claim 2, The one or more processors present the plurality of options to the operator such that the recommended options are more prominent than options other than the recommended options. Remote support system.

4. A remote assistance system according to any one of claims 1 to 3, The plurality of vehicle processes include: A recognition process for recognizing a situation around the autonomous vehicle; an action determination process for determining whether or not to execute an action based on a result of the recognition process; a timing determination process for determining an execution timing for executing the action; Includes Remote support system.

5. A remote assistance method for remotely assisting driving of an autonomous vehicle, comprising: The autonomous vehicle includes: A plurality of vehicle processes including recognition and decision making during autonomous driving; an assistance request process for sending an assistance request for requesting assistance for at least one of the plurality of vehicle processes to a remote assistance system that remotely supports driving of the autonomous vehicle; Run The performance database indicates a performance of the assistance request processing by the autonomous driving vehicle for each scene, The remote assistance method includes: an operator instruction request process that requests an operator to provide instructions for a target autonomous vehicle in a first scene in response to the assistance request sent from the target autonomous vehicle in a first scene; an operator instruction notification process for notifying the target autonomous driving vehicle of the instruction from the operator; Including, The assistance process is a vehicle process among the plurality of vehicle processes in the first scene that is performed by the operator instead of the target autonomous vehicle, A delegated process is a vehicle process among the plurality of vehicle processes in the first scene that the operator entrusts to the target autonomous vehicle, The target autonomous vehicle executes the delegation process without relying on instructions from the operator; and The operator instruction request process includes: obtaining the performance of the assistance request process related to the first scene from the performance database; determining a recommended option to be recommended to the operator from among a plurality of options of combinations of the support process and the delegation process based on the performance of the support request process related to the first scene; A process of presenting at least the recommended options to the operator; Includes Remote assistance methods.

Images