JP2022095350A - Remote support device, remote support method, and remote support program - Google Patents

Abstract

To provide a remote support device, a remote support method, and a remote support program capable of improving accuracy of an automatic response when performing remote support for an automatic drive vehicle.SOLUTION: A remote support device (20) includes: a request acquisition unit (21A) for acquiring support request information including position information of an automatic drive vehicle from the automatic drive vehicle, and an automatic determination unit (21B). The automatic determination unit acquires a determination scene immediately responded by an operator in a specific area including a point indicated by position information from an operator determination log DB (25B) according to the acquired support request information, classifies a scene of the automatic drive vehicle into either scene to overtake or scene to wait based on the acquired determination scene to set an effective period representing a period for outputting an automatic response based on an immediate response of the operator as a valid response when the scene of the automatic drive vehicle is classified into the scene to overtake shorter than when it is classified into the scene to wait.SELECTED DRAWING: Figure 3

Description

The present disclosure relates to remote support devices, remote support methods, and remote support programs.

There is a technology to remotely support autonomous vehicles from the autonomous driving support center. For example, when a running autonomous vehicle stops due to some kind of obstacle, a support request is made to the autonomous driving support center. The operator of the autonomous driving support center determines a scene such as parking on the road or construction in response to a support request, and makes a response to the autonomous driving vehicle based on the determination scene. After that, an automatic response based on the operator's response is performed in response to the support request to the autonomous driving vehicle traveling at the point where the support request is made.

For example, Patent Document 1 describes a control device for remotely controlling an autonomous driving vehicle. The control device is point data including position information acquired from the target vehicle, which is the autonomous driving vehicle subject to remote control, and sensing data collected by the sensors of the target vehicle when remote control is performed. It has a registration unit to register. In addition, when the control device executes remote control, if point data similar to the current point data acquired from the target vehicle is available among the point data registered by the registration unit, the control device uses similar point data. Select the first remote control to instruct driving by remote control using it, and when similar point data is not available, accept the command input to the interface by the operator and instruct the driving by remote control based on the command. A selection unit for selecting a second remote control to be performed is provided.

Japanese Unexamined Patent Publication No. 2019-185280

In the technique described in Patent Document 1, when the point data are similar, the same automatic response is performed. However, there are cases where the validity period is not specified as to how long the automatic response should be continued. In this case, the same automatic response as before may occur and the autonomous vehicle may behave incorrectly, even though another failure has actually occurred. Therefore, it is desired to improve the accuracy of the automatic response.

It is an object of the present disclosure to provide a remote support device, a remote support method, and a remote support program that can improve the accuracy of automatic response when performing remote support for an autonomous driving vehicle.

In order to achieve the above object, the remote support device (20) according to the first aspect of the present disclosure is a remote support device that provides remote support to an autonomous driving vehicle in a controlled area by an autonomous driving support center. , The acquisition unit (21A) that acquires the support request information including the position information of the autonomous driving vehicle from the autonomous driving vehicle, and the determination that the operator has responded in the past according to the support request information acquired by the acquisition unit. From the database (25B) including the operator judgment log which is a log of the scene, the judgment scene most recently responded by the operator in a specific area including the point indicated by the position information is acquired, and based on the acquired judgment scene. , The classification unit (21B) that classifies the scene of the autonomous driving vehicle into either a scene to be overtaken or a scene to be on standby, and the classification unit classifies the scene of the autonomous driving vehicle into the scene to be overtaken. In this case, the setting unit (21B) sets the valid period indicating the period for outputting the automatic response based on the operator's latest response as a valid response shorter than that when classified into the scene to be waited. It is equipped with.

The remote support method according to the second aspect of the present disclosure is a remote support method using a remote support device that provides remote support to an automatically driving vehicle in a controlled area by the automatic driving support center, from the automatic driving vehicle. The position information is acquired from a database including an operator determination log, which is a log of a determination scene in which the operator has responded in the past in response to the acquired support request information by acquiring support request information including the position information of the autonomous driving vehicle. The judgment scene most recently responded by the operator in the specific area including the point indicated by the operator is acquired, and based on the acquired determination scene, the scene of the autonomous driving vehicle is overtaken and the scene to be waited. When the scene of the self-driving vehicle is classified into any of the scenes to be overtaken, the valid period representing the period for outputting the automatic response based on the operator's latest response as a valid response is defined as the above. Set it shorter than when it is classified as a scene to wait.

The remote support program (25A) according to the third aspect of the present disclosure is a remote support program of a remote support device that provides remote support to an automatically driven vehicle in a managed area by an automatic driving support center, and is a remote support program for a computer. The acquisition unit (21A) that acquires the support request information including the position information of the automatic driving vehicle from the automatic driving vehicle, and the determination scene in which the operator has responded in the past in response to the support request information acquired by the acquisition unit. From the database (25B) including the operator judgment log which is a log, the judgment scene most recently responded by the operator in a specific area including the point indicated by the position information is acquired, and based on the acquired judgment scene, the said The classification unit (21B) that classifies the scene of the autonomous vehicle into either a scene to be overtaken or a scene to be on standby, and the classification unit classifies the scene of the autonomous vehicle into the scene to be overtaken. In this case, it functions as a setting unit (21B) that sets a valid period indicating a period for outputting an automatic response based on the operator's latest response as a valid response shorter than that when classified into the scene to be waited for. Let me.

According to the disclosed technique, there is an effect that the accuracy of the automatic response when performing remote support for the autonomous driving vehicle can be improved.

It is a figure which shows an example of the structure of the remote support system which concerns on 1st Embodiment. It is a figure which provides the explanation of the remote support by the operator which concerns on embodiment. It is a block diagram which shows an example of the functional configuration of the in-vehicle device and the remote support device which concerns on 1st Embodiment. It is a figure which shows an example of the scene classification table which concerns on embodiment. It is a figure which provides the explanation of the validity period setting process which concerns on 1st Embodiment. It is a figure which shows an example of the duration distribution for each determination scene which concerns on embodiment. It is a figure which provides the explanation of the calculation method of the duration time which concerns on embodiment. It is a figure which provides the explanation of another calculation method of the duration length which concerns on embodiment. It is a figure which provides the explanation of the method of the same determination which concerns on embodiment. It is a flowchart which shows an example of the processing flow by the remote support program which concerns on 1st Embodiment. It is a figure which provides the explanation of the validity period setting process which concerns on the comparative example. It is a figure which shows an example of the automatic response non-response period which concerns on 2nd Embodiment. It is a figure which provides the explanation of the valid period setting process which concerns on 2nd Embodiment. FIGS. (A) to (C) are diagrams for explaining a method of calculating the duration distribution for each place associated with the determination scene according to the fourth embodiment. It is a figure which shows an example of the duration model with respect to the distance from a facility. It is a figure which provides the explanation of the method of calculating the continuation time distribution for each time zone associated with the determination scene which concerns on 4th Embodiment. It is a figure which shows an example of the duration model for precipitation. It is a figure which provides the explanation of the method of calculating the duration distribution for each vehicle type associated with the determination scene which concerns on 4th Embodiment. It is a block diagram which shows an example of the functional structure of the vehicle-mounted device and the remote support device which concerns on 5th Embodiment.

Hereinafter, an example of a mode for carrying out the technique of the present disclosure will be described in detail with reference to the drawings.

[First Embodiment]
FIG. 1 is a diagram showing an example of the configuration of the remote support system 100 according to the first embodiment.

As shown in FIG. 1, the remote support system 100 according to the present embodiment includes an in-vehicle device 10 mounted on an autonomous driving vehicle and a remote support device 20 provided in an autonomous driving support center. The remote support device 20 provides remote support to the autonomous driving vehicle in the area managed by the autonomous driving support center. In this embodiment, a privately-owned passenger car will be described as an example of an autonomous driving vehicle, but it may be applied to other vehicles such as trucks, buses, and taxis, for example. The self-driving vehicle also includes the case of a manned person who rides to control the vehicle or to replace the control of the vehicle in an emergency. Further, the vehicle includes a vehicle in which a part of the steering of the vehicle is automatically performed.

The in-vehicle device 10 and the remote support device 20 are communicably connected via the network N. As an example, the Internet, WAN (Wide Area Network), and the like are applied to the network N.

An autonomous driving vehicle is a vehicle capable of autonomous driving under predetermined conditions regardless of the driver's operation. When any event such as parking on the road, traffic jam, construction, etc. occurs while driving, the autonomous driving vehicle performs an operation of overtaking or waiting, and these operations are performed based on the response from the autonomous driving support center as an example. Will be.

The in-vehicle device 10 has a function of generating a travel plan including a travel route to the destination based on information of the destination such as an address or latitude and longitude, and a function of controlling automatic driving of the own vehicle. The in-vehicle device 10 includes a CPU (Central Processing Unit) 11, a memory 12, a display unit 13, a storage unit 14, a sensor group 15, a camera 16, and a communication unit 17.

The CPU 11 is an example of a processor. The term "processor" as used herein refers to a processor in a broad sense, and is a general-purpose processor (for example, CPU) or a dedicated processor (for example, GPU: Graphics Processing Unit, ASIC: Application Specific Integrated Circuit, FPGA: Field Programmable Gate Array). , Programmable logic devices, etc.). The memory 12 is composed of a ROM (Read Only Memory), a RAM (Random Access Memory), and the like.

For the display unit 13, for example, a liquid crystal display (LCD), an organic EL (Electro Luminescence) display, or the like is used. The display unit 13 may have a touch panel integrally.

For the storage unit 14, for example, an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like is used. A control program (not shown) for controlling automatic operation is stored in the storage unit 14.

The sensor group 15 is composed of various sensors for grasping the situation around the own vehicle. The sensor group 15 includes a millimeter-wave radar that transmits exploration waves to a predetermined range outside the vehicle, and LIDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging) that scans at least a predetermined range in front of the vehicle. Further, the sensor group 15 may include a GNSS (Global Navigation Satellite System) receiver mounted on the own vehicle. Information such as the current position and the current time of the own vehicle is acquired by this GNSS receiver.

The camera 16 captures a predetermined range in a predetermined direction of the own vehicle. Specifically, the camera 16 is provided on the front side of the own vehicle, and photographs the front area of the own vehicle. The number of cameras 16 may be one, but a plurality of cameras may be provided at a plurality of locations in order to obtain more information.

The communication unit 17 is a communication interface for connecting to a network N such as the Internet or WAN and communicating with the remote support device 20.

The in-vehicle device 10 is connected to a traveling device (not shown) necessary for automatic driving, and automatic driving is performed by controlling this traveling device. The traveling device includes, for example, an electric power steering, an electronically controlled brake, an electronically controlled throttle, and the like.

The in-vehicle device 10 performs automatic driving by controlling driving, steering, and braking of the own vehicle so as to automatically drive according to the traveling plan of the own vehicle. There are various known methods for the automatic operation method itself, and the present embodiment is not particularly limited.

On the other hand, the remote support device 20 monitors the vehicle state of the autonomous driving vehicle by periodically communicating with the in-vehicle device 10 of the autonomous driving vehicle. As an example, a general-purpose computer device such as a server computer or a personal computer (PC) is applied to the remote support device 20. The remote support device 20 includes a CPU 21, a memory 22, an operation unit 23, a display unit 24, a storage unit 25, and a communication unit 26.

The CPU 21 is an example of a processor. As described above, the term processor here refers to a processor in a broad sense, and includes a general-purpose processor and a dedicated processor. The memory 22 is composed of a ROM, a RAM, and the like.

The operation unit 23 is configured as an interface for receiving an operation input to the remote support device 20. For the display unit 24, for example, a liquid crystal display (LCD), an organic EL display, or the like is used. The display unit 24 may have a touch panel integrally.

For the storage unit 25, for example, an HDD, SSD, flash memory, or the like is used. The storage unit 25 stores a remote support program 25A for remotely supporting an autonomous driving vehicle. The remote support program 25A may be pre-installed in, for example, the remote support device 20. The remote support program 25A may be realized by storing it in a non-transitory non-transitory recording medium or distributing it via the network N and appropriately installing it in the remote support device 20. Examples of non-volatile non-transitional recording media include CD-ROMs (Compact Disc Read Only Memory), optomagnetic disks, HDDs, DVD-ROMs (Digital Versatile Disc Read Only Memory), flash memories, memory cards, and the like. Is assumed.

The communication unit 26 is a communication interface for connecting to a network N such as the Internet or WAN and communicating with the in-vehicle device 10.

FIG. 2 is a diagram for explaining remote support by an operator according to the present embodiment.

As shown in FIG. 2, when the running automatic driving vehicle 30 is stopped, a support request is made to the automatic driving support center. The operator OP of the automatic driving support center determines a scene such as parking on the road or construction in response to a support request from the automatic driving vehicle 30, and makes a response to the automatic driving vehicle 30 based on the determination scene. At this time, the operator OP registers the responded determination scene in association with the point data on the map. After that, an automatic response based on the same determination scene is performed in response to a support request to the autonomous driving vehicle traveling at the point where the determination scene is registered.

By the way, as described above, there are cases where the validity period is not defined as to how long the automatic response should be continued. In this case, even though another failure has actually occurred, the same automatic response as before is performed, and the autonomous driving vehicle 30 may behave erroneously. Therefore, it is desired to improve the accuracy of the automatic response.

Therefore, the CPU 11 of the in-vehicle device 10 according to the present embodiment functions as each unit shown in FIG. 3 by writing the control program stored in the storage unit 14 to the RAM and executing the control program. Further, the CPU 21 of the remote support device 20 according to the present embodiment functions as each unit shown in FIG. 3 by writing the remote support program 25A stored in the storage unit 25 into the RAM and executing the remote support program 25A.

FIG. 3 is a block diagram showing an example of the functional configuration of the in-vehicle device 10 and the remote support device 20 according to the first embodiment.

As shown in FIG. 3, the CPU 11 of the vehicle-mounted device 10 according to the present embodiment functions as a request output unit 11A and a determination acquisition unit 11B.

The request output unit 11A sends support request information including time information, position information, sensing information, etc. when it is determined that the own vehicle cannot continue traveling to the remote support device 20 of the automatic driving support center via the communication unit 17. Output (send) to.

The determination acquisition unit 11B acquires (receives) the operator response or automatic response output from the remote support device 20 via the communication unit 17, and captures the determination scene included in the acquired operator response or automatic response in the vehicle of the own vehicle. Send to the control device (not shown) that controls.

On the other hand, the CPU 21 of the remote support device 20 according to the present embodiment functions as a request acquisition unit 21A, an automatic determination unit 21B, a determination output unit 21C, an operator presentation unit 21D, a duration calculation unit 21E, and an operator determination registration unit 21F. .. The request acquisition unit 21A is an example of an acquisition unit, the automatic determination unit 21B is an example of a classification unit, a setting unit, and an instruction unit, and the duration calculation unit 21E is an example of a calculation unit.

In the storage unit 25, for example, an operator determination log database (hereinafter referred to as “operator determination log DB”) 25B and a duration distribution database (hereinafter referred to as “duration time distribution DB”) 25C are stored. The operator determination log DB 25B and the duration distribution DB 25C may be stored in an external storage device instead of the storage unit 25.

The operator determination log is registered in the operator determination log DB 25B. This operator determination log is a log of determination scenes that the operator has responded to in the past. This determination scene represents the result of the operator determining the current scene of the autonomous driving vehicle as, for example, "parking on the road", "construction", "traffic jam", etc., according to the support request information from the autonomous driving vehicle. ing. Each operator judgment log contains, for example, an ID (Identification) unique to each log, a judgment time, a support request position (or a support request area), support contents (including judgment scenes, control contents, etc.), and sensing information (peripheral image). , Point cloud, recognition result, etc.), presence / absence of the same judgment (including the ID of the same operator judgment log if present), and the like. Further, the duration distribution is registered in the duration distribution DB 25C. This duration distribution includes, for example, information such as scene information about a scene which is a unit for managing the duration, a duration distribution (for example, a histogram), and the like. The calculation method of this duration distribution will be described later.

The request acquisition unit 21A acquires (receives) the support request information output from the autonomous driving vehicle via the communication unit 26, and sends the acquired support request information to the automatic determination unit 21B.

The automatic determination unit 21B responded most recently by the operator from the operator determination log DB 25B in a specific area including the point indicated by the position information of the autonomous driving vehicle in response to the support request information acquired by the request acquisition unit 21A. Acquire the judgment scene. Here, the specific region is defined as, for example, a circular region having a diameter of X m centered on a support request point.

Based on the acquired determination scene, the automatic determination unit 21B classifies the scene of the autonomous driving vehicle into one of a scene to be overtaken and a scene to be on standby.

FIG. 4 is a diagram showing an example of the scene classification table 25D according to the present embodiment.

The scene classification table 25D shown in FIG. 4 is stored in the storage unit 25, for example, and can be referred to by the automatic determination unit 21B. If the judgment scene is, for example, "parking on the road", "construction", "accident", etc., the scene of the autonomous driving vehicle is classified into the scene to be overtaken, and the judgment scene is, for example, "pedestrian", "signal". If it is "waiting", "traffic jam", "waiting for a bus", etc., the scene of the autonomous driving vehicle is classified as a scene to be waited.

However, in the case of a scene to be overtaken, there is a problem of erroneous determination that an unnecessary overtaking occurs if the scene is actually classified as a scene to be overtaken even though it is a scene to be waited for. On the other hand, in the case of a scene that should be waited, there is a problem of erroneous determination that waiting occurs if the scene is actually classified as a scene that should be waited even though it is a scene that should be overtaken.

The automatic determination unit 21B sets the valid period shorter when the scene of the autonomous driving vehicle is classified into the scene to be overtaken than when it is classified into the scene to be waited. The valid period represents a period during which an automatic response based on the operator's latest response is output as a valid response. The validity period is derived from the latest operator response.

FIG. 5 is a diagram for explaining the validity period setting process according to the first embodiment.

As shown in FIG. 5, in the graph 40, the horizontal axis indicates the duration, and the vertical axis indicates the frequency and the existence probability. Graph 40 shows, as an example, the case of a “street parking” scene. The existence probability 41 indicates the probability that the “street parking” scene continues to exist, and the probability decreases with the passage of time. Frequency 42 indicates the frequency with which the “street parking” scene occurs for each duration zone. For example, parking for 5 minutes or more and 10 minutes or less occurs 10 times. The duration distribution 43 is calculated for each determination scene, for example, from the operator determination log included in the operator determination log DB 25B. The duration distribution 43 represents the relationship between the frequency of occurrence of the determination scene and the duration of the determination scene. The duration distribution 43 is calculated by the duration calculation unit 21E, which will be described later.

As an example, as shown in the graph 40 of FIG. 5, the automatic determination unit 21B sets the effective period when the scenes to be waited are the average value μ of the duration distribution 43, and the scenes to be overtaken. The validity period t in the case of classification is set to be shorter than the average value μ, that is, the existence probability 41 is set to be relatively high. The validity period t is calculated by, for example, the following formula (1). However, μ indicates the average value of the duration distribution 43, and σ indicates the standard deviation of the duration distribution 43.

t = μ-3σ ・ ・ ・ (1)

In (S1) of FIG. 5, since the autonomous driving vehicle 30 encounters some kind of obstacle while traveling, it temporarily stops and requests support from the autonomous driving support center. The operator (OP) of the autonomous driving support center determines that the scene encountered by the autonomous driving vehicle 30 is "parking on the road", and responds with the determination scene which is the determination result. The self-driving vehicle 30 determines that the vehicle is overtaking according to the operator's response, and performs an operation of overtaking the vehicle "parked on the road".

Next, in (S2), similarly to the above (S1), the autonomous driving vehicle 30 encounters some kind of obstacle, so that the autonomous driving vehicle 30 is temporarily stopped and a support request is made to the autonomous driving support center. The remote support device 20 of the automatic driving support center automatically responds to "parking on the road" as an automatic response based on the latest operator response. The self-driving vehicle 30 determines that the vehicle is overtaking according to the automatic response, and performs an operation of overtaking the vehicle "parked on the road".

Next, in (S3), similarly to the above (S2), the autonomous driving vehicle 30 encounters some kind of obstacle, so that the autonomous driving vehicle 30 is temporarily stopped and a support request is made to the autonomous driving support center. The remote support device 20 of the automatic driving support center automatically responds to "parking on the road" as an automatic response based on the latest operator response. The self-driving vehicle 30 determines that the overtaking scene is in accordance with the automatic response, and performs an overtaking operation. However, at this point, there is no vehicle for "parking on the street", and there is a possibility that the scene has changed to a waiting scene such as "pedestrian" or "waiting for a traffic light". In other words, it can be said that it is not desirable from the viewpoint of safety to perform an overtaking operation even though the scene should actually be on standby.

Therefore, in the present embodiment, as described above, the valid period t of the scene to be overtaken is set shorter than the valid period t of the scene to be waited (for example, the average value μ). In other words, since the scene that should be on standby only waits, it can be said that there is no problem from the viewpoint of safety even if the scene changes to the scene that should be overtaken within the valid period. On the other hand, in the scene to be overtaken, if the scene changes to a scene to wait within the valid period, as described above, it may not be desirable from the viewpoint of safety. Therefore, the valid period t of the scene to be overtaken is set short.

Further, when the period from the latest operator response to the acquisition of the support request information is within the valid period, the automatic determination unit 21B instructs the determination output unit 21C to output the automatic response to the autonomous driving vehicle. .. On the other hand, if the period from the latest operator response to the acquisition of the support request information exceeds the valid period, the automatic determination unit 21B stops the automatic response, sends the support request information to the operator presentation unit 21D, and sends the support request information to the operator this time. Instruct to register the judgment scene.

The operator presentation unit 21D presents the forward image, map information, sensing information, etc. included in the support request information to the operator, and accepts the registration of the current determination scene from the operator. The operator presentation unit 21D instructs the determination output unit 21C to output the operator response to the autonomous driving vehicle.

The determination output unit 21C outputs (transmits) the automatic response as a determination result according to the output instruction of the automatic response from the automatic determination unit 21B, or determines the operator response according to the output instruction of the operator response from the operator presentation unit 21D. Output as a result.

Here, as described above, the duration calculation unit 21E calculates the duration distribution for each determination scene using the operator determination log included in the operator determination log DB 25B, and distributes the calculated duration distribution to the duration distribution. Register in DB25C. The automatic determination unit 21B sets the valid period according to the scene of the autonomous driving vehicle based on the duration distribution registered in the duration distribution DB 25C.

FIG. 6 is a diagram showing an example of the duration distribution for each determination scene according to the present embodiment.

As shown in FIG. 6, a different duration distribution may be calculated for each determination scene such as “parking on the street” or “obstacle”. In this case, it is possible to grasp a different duration distribution for each determination scene. Therefore, the accuracy of the validity period can be further improved.

FIG. 7 is a diagram for explaining a method of calculating the duration length according to the present embodiment.

As an example, the continuation time calculation unit 21E calculates the continuation time length when the determination of the same scene is continuously performed from the operator determination log acquired from the operator determination log DB 25B, as shown in FIG. That is, the difference between the time when the operator judgment is made first (first time) and the time when the same operator judgment is made last (here, the third time) is defined as the duration length (same judgment continuation time). calculate.

FIG. 8 is a diagram for explaining another calculation method of the duration length according to the present embodiment.

In FIG. 8, x indicates a travel log. As shown in FIG. 8, the duration calculation unit 21E may calculate the duration length using the travel log of the autonomous driving vehicle. That is, based on the duration length (same judgment duration) calculated from the operator judgment log, several tens of seconds (for example, 20 seconds) before and after are set in advance as a predetermined period, and the running log and infrastructure are installed in the predetermined period. Get camera logs etc. Then, the reference duration length is corrected by the period in which the sensing information obtained from these logs is similar. As a result, the duration length can be grasped more accurately.

Further, the operator judgment registration unit 21F determines whether or not the latest operator judgment log and the current operator judgment result are the same judgment scene, and the current operator judgment result is combined with the presence or absence of the same judgment. , Registered in the operator determination log DB 25B.

FIG. 9 is a diagram for explaining the method of the same determination according to the present embodiment.

The operator determination registration unit 21F acquires an operator determination log, which is a log of the operator determination most recently performed in the same area as the operator determination result this time, from the operator determination log DB 25B. Then, as an example, the operator determination registration unit 21F determines whether or not the current operator determination result and the latest operator determination log in the same area are the same determination scene, as shown in FIG. If both determination scenes are, for example, "parking on the street", it is determined that they are the same. For the determination of whether or not they are the same, the operator may be prompted to make a judgment by presenting a front image or the like obtained from each autonomous driving vehicle to the operator. Alternatively, it may be mechanically determined by using a known method whether or not the feature quantities of the sensing information obtained from each autonomous driving vehicle are similar.

As shown in FIG. 3 above, the remote support device 20 according to the present embodiment calculates the duration of the determination scene by the past operator in advance, and registers the calculated duration distribution in the duration distribution DB 25C. Then, when the validity period is set, the duration distribution is read from the duration distribution DB 25C, and the validity period is set based on the read duration distribution.

Next, with reference to FIG. 10, the operation of the remote support device 20 according to the first embodiment will be described.

FIG. 10 is a flowchart showing an example of the processing flow by the remote support program 25A according to the first embodiment.

First, when the power of the remote support device 20 is turned on, the remote support program 25A is started, and each of the following steps is executed.

In step S101 of FIG. 10, it is determined whether or not the CPU 21 has acquired the support request information from the in-vehicle device 10 of the autonomous driving vehicle. When it is determined that the support request information has been acquired from the in-vehicle device 10 (in the case of affirmative determination), the process proceeds to step S102, and when it is determined that the support request information has not been acquired from the in-vehicle device 10 (in the case of a negative determination). It becomes a standby in step S101. At this time, based on the position information included in the acquired support request information, the point of the autonomous driving vehicle and the area including the point (for example, a circular area having a diameter of X m around the point) are specified.

In step S102, the CPU 21 acquires the latest operator determination log in the same area as the area specified in step S101 from the operator determination log DB 25B. This operator determination log includes the operator's latest determination scene.

In step S103, the CPU 21 classifies the scenes of the autonomous driving vehicle by referring to the scene classification table 25D shown in FIG. 4 above as an example based on the judgment scene included in the latest operator judgment log acquired in step S102. do.

In step S104, the CPU 21 determines whether the scene of the autonomous driving vehicle is classified into a scene to be overtaken or a scene to be on standby as a result of the classification in step S103. If it is determined that the scene should be overtaken (in the case of a scene to be overtaken), the process proceeds to step S105, and if it is determined that the scene is classified into a scene to be waited in (in the case of a scene to be waited), the process proceeds to step S106. do.

In step S105, the CPU 21 sets the valid period t of the scene to be overtaken to be shorter than the valid period t of the scene to be waited for (for example, the average value μ). The validity period t of the scene to be overtaken is calculated using, for example, the above equation (1).

In step S106, the CPU 21 sets the average value μ of the duration distribution as an example, as shown in FIG. 5 above, as the valid period of the scene to be waited for.

In step S107, the CPU 21 determines whether or not the period from the latest operator response to the acquisition of support request information is within the valid period (here, the valid period t or the average value μ). If it is determined that the validity period is within the valid period (in the case of affirmative determination), the process proceeds to step S108, and if it is determined that the valid period is exceeded (in the case of a negative determination), the process proceeds to step S109.

In step S108, the CPU 21 instructs to output an automatic response.

In step S109, the CPU 21 stops the automatic response and instructs the operator to register the current determination scene.

In step S110, the CPU 21 outputs an automatic response based on the instruction in step S108 or an operator response based on the instruction in step S109 to the autonomous driving vehicle.

In step S111, the CPU 21 determines whether or not the end timing has arrived, for example, a remote support end instruction or the like. When it is determined that the end timing has not arrived (in the case of a negative determination), the return process is repeated in step S101, and when it is determined that the end timing has arrived (in the case of an affirmative determination), a series of processes by the remote support program 25A is performed. finish.

As described above, according to the present embodiment, the valid period of the scene to be overtaken is set shorter than the valid period of the scene to be waited for. Therefore, the accuracy of the automatic response when performing remote support for the autonomous driving vehicle is improved.

[Second Embodiment]
In the first embodiment, the embodiment in which the valid period of the scene to be overtaken is set shorter than the valid period of the scene to be waited for has been described. In the present embodiment, only in the case of a scene where unnecessary overtaking is not desirable, a mode in which the valid period of the scene to be overtaken is set shorter than the valid period of the scene to be waited for will be described.

The remote support device according to the present embodiment has the same components as the remote support device 20 described in the first embodiment, and only the differences will be described with reference to FIG. 3 described above.

First, with reference to FIG. 11, a case where the effective period t of the scene to be overtaken is set to the average value μ of the duration distribution 43 will be described.

FIG. 11 is a diagram for explaining the validity period setting process according to the comparative example.

In the comparative example of FIG. 11, the effective period t of the “street parking” scene, which is a scene to be overtaken, is set to the average value μ of the duration distribution 43. Also, the actual duration of the "street parking" scene is shorter than the validity period t.

In FIG. 11 (S11), since the autonomous driving vehicle 30 encounters some kind of obstacle while traveling, it temporarily stops and requests support from the autonomous driving support center. The operator (OP) of the autonomous driving support center determines that the scene encountered by the autonomous driving vehicle 30 is "parking on the road", and responds with the determination scene which is the determination result. The self-driving vehicle 30 determines that the vehicle is overtaking according to the operator's response, and performs an operation of overtaking the vehicle "parked on the road".

Next, in (S12), similarly to the above (S11), the autonomous driving vehicle 30 encounters some kind of obstacle, so that the autonomous driving vehicle 30 is temporarily stopped and a support request is made to the autonomous driving support center. The remote support device 20 of the automatic driving support center automatically responds to "parking on the road" as an automatic response based on the latest operator response. The self-driving vehicle 30 determines that the overtaking scene is in accordance with the automatic response, and performs an overtaking operation. However, at this point, there is no vehicle "parked on the street", and "traffic jam", which is a scene to be waited for, has occurred, so that the automatic response determination scene is an erroneous determination. The period T1 represents a period in which an erroneous determination is made.

As shown in FIG. 11, the erroneous determination A shows a case where "traffic jam" is erroneously determined as "parking on the road" at a place of two lanes on each side. In this case, there is a problem that unnecessary overtaking occurs in the adjacent lane. On the other hand, the erroneous determination B shows a case where "traffic jam" is erroneously determined as "parking on the road" at a place of one lane on each side. In this case, there is a problem that unnecessary overtaking occurs in the oncoming lane.

Therefore, in the valid period t of the scene to be overtaken, the specific area for which the support request is made is the area where the occurrence frequency of the scene to be waited is equal to or higher than the preset threshold value, or the occurrence frequency of the scene to be waited is set. When the region is higher than the frequency of occurrence of the scenes to be overtaken, the above equation (1) may be applied and set shorter than the valid period (for example, the average value μ) of the scenes to be waited for. The threshold value is, for example, a value experimentally obtained in advance, and as an example, 0.5 or the like is set when the occurrence frequency is expressed in the range of 0 or more and 1 or less. The occurrence frequency f of a certain scene A in the same region is calculated by, for example, the following equation (2). However, N indicates the number of times the scene A is determined in the same area, and R indicates the number of times the autonomous driving vehicle has traveled in the same area.

f = N / R ・ ・ ・ (2)

The occurrence frequency f may be calculated for each major classification of scenes (scenes to be waited for, scenes to be overtaken), or for each time zone.

That is, in a place where traffic congestion is likely to occur on a daily basis, it is possible to suppress an erroneous determination due to an automatic response by shortening the effective period t of a scene to be overtaken such as "parking on the street".

FIG. 12 is a diagram showing an example of the automatic response non-response period T2 according to the second embodiment.

As shown in FIG. 12, when the valid period t is set shorter than the valid period (for example, the average value μ) of the scene to be waited for the scene to be overtaken such as “parking on the street”, unnecessary overtaking is performed. Can be suppressed. However, on the other hand, when the duration of the scene to be overtaken is actually longer than the valid period t, the automatic response non-response period T2, which is the period during which the automatic response cannot be performed, occurs. In the present embodiment, it is considered that the period during which the automatic response is not possible T2 is shortened as much as possible and the period during which the automatic response is possible is lengthened.

For example, when the specific area for which support is requested is an area where a scene to be overtaken such as "parking on the street" is likely to occur, setting a long validity period t does not often cause a big safety problem.

Therefore, in the automatic determination unit 21B according to the present embodiment, the specific area for which the support request is made is a region in which the occurrence frequency of the scene to be overtaken is equal to or higher than a preset threshold value, or the occurrence frequency of the scene to be overtaken. When is in a region higher than the frequency of occurrence of the scene to be waited for, the valid period t of the scene to be overtaken is set to be equal to or longer than the valid period t of the scene to be waited (for example, the average value μ). The threshold value is, for example, a value experimentally obtained in advance. In other words, by deciding whether or not to set the validity period t short according to the characteristics of the area where the scene to be overtaken occurs, while securing a certain period during which automatic response is possible, erroneous judgment due to automatic response is suppressed. do.

FIG. 13 is a diagram for explaining the validity period setting process according to the second embodiment.

In FIG. 13, the black circle on the map indicates the point where the determination of "parking on the street" is registered. From the graph showing the frequency of occurrence of each scene that occurred in the past for point A, the frequency of occurrence of "traffic jam" is higher than the frequency of occurrence of "parking on the street" (or the frequency of occurrence of "traffic jam" is preset. It is above the threshold). In this case, if the valid period t of "street parking" is relatively long, there is a high possibility that the "traffic jam" will change within the valid period t, and there is a high possibility that the automatic response will be an erroneous determination. Therefore, the effective period t is set shorter than the average value μ. On the other hand, at point B, from the graph showing the frequency of occurrence of each scene that occurred in the past, the frequency of occurrence of "street parking" is higher than the frequency of occurrence of "traffic jam" (or the frequency of occurrence of "street parking" is in advance. It is above the set threshold value). In this case, even if the valid period t of "street parking" is relatively long, it is unlikely that the vehicle will change to "traffic jam" within the valid period t, and it is unlikely that the automatic response will be erroneously determined. Therefore, the effective period t is set to the average value μ.

As described above, according to the present embodiment, it is determined whether or not the validity period is set short depending on the characteristics of the region where the scene to be overtaken is generated. Therefore, while ensuring a certain period during which automatic response is possible, erroneous determination due to automatic response is suppressed.

[Third Embodiment]
In the first embodiment, the mode of classifying the scenes of the autonomous driving vehicle based on the latest operator's judgment scene in a specific area obtained from the operator judgment log has been described. In the present embodiment, in addition to the determination scene, a mode for classifying the scene of the autonomous driving vehicle by using the similarity of the sensing information will be described.

The remote support device according to the present embodiment has the same components as the remote support device 20 described in the first embodiment, and only the differences will be described with reference to FIG. 3 described above.

The operator determination log registered in the operator determination log DB 25B includes sensing information (hereinafter, referred to as “first sensing information”) obtained from the latest autonomous driving vehicle that has received the operator's latest response. Further, the support request information includes the sensing information (hereinafter referred to as "second sensing information") obtained from the autonomous driving vehicle this time.

The automatic determination unit 21B classifies the scene of the autonomous driving vehicle into either a scene to be overtaken or a scene to be on standby based on the degree of similarity between the first sensing information and the second sensing information. A known method is used to calculate the degree of similarity. For example, when an image is used as the first sensing information and the second sensing information, pattern matching or the like is applied as an example.

The automatic determination unit 21B has a remaining valid period of a certain period (for example, 20 seconds) or more for the period from the latest response by the operator to the acquisition of support request information, and the similarity is the first threshold value (for example). , 0.6) or more, give an instruction to output an automatic response to the autonomous driving vehicle. On the other hand, in the automatic determination unit 21B, the remaining time of the valid period for the period from the latest response by the operator to the acquisition of support request information is a certain time (for example, 20 seconds) or more, and the similarity is the first threshold value. If it is less than (for example, 0.6), the automatic response is stopped and the operator is instructed to register the current determination scene. Further, the automatic determination unit 21B has a remaining valid period of less than a certain time (for example, 20 seconds) with respect to the period from the latest response by the operator to the acquisition of support request information, and the similarity is the first threshold value. If it is greater than or equal to a second threshold value (for example, 0.9), an instruction is given to output an automatic response to the autonomous driving vehicle. On the other hand, the automatic determination unit 21B has a remaining valid period of less than a certain time (for example, 20 seconds) with respect to the period from the latest response by the operator to the acquisition of support request information, and the similarity is the second threshold value. If it is less than (for example, 0.9), the automatic response is stopped and the operator is instructed to register the current determination scene. The valid period referred to here may be the valid period t of the scene to be overtaken or the valid period of the scene to be waited for (for example, the average value μ).

Further, when the similarity is equal to or higher than the threshold value (for example, 0.6), the automatic determination unit 21B automatically performs an automatic response if the period from the latest response by the operator to the acquisition of support request information is within the valid period. When an instruction is given to output to the driving vehicle and the similarity is less than the threshold value (for example, 0.6), an automatic response is made even if the period from the latest response by the operator to the acquisition of support request information is within the valid period. May be canceled and the operator may be instructed to register the judgment scene this time. The valid period referred to here may be the valid period t of the scene to be overtaken or the valid period of the scene to be waited for (for example, the average value μ), as described above.

The above threshold value may be set to a larger value as the remaining time of the valid period with respect to the period from the latest response by the operator to the acquisition of support request information becomes smaller. As a result, when the remaining time of the valid period is long, the similarity determination is relatively loose, and when the remaining time of the valid period is short, the similarity determination becomes relatively strict.

As described above, according to the present embodiment, the scenes of the autonomous driving vehicle are classified by using the similarity of the sensing information in addition to the determination scenes. Therefore, the scenes of self-driving cars are classified with high accuracy.

[Fourth Embodiment]
In the first embodiment described above, a mode in which the determination scene is used as a unit for managing the duration distribution has been described. In this embodiment, a mode in which a place, a time zone, and a vehicle type are used as a unit for managing the duration distribution will be described.

The remote support device according to the present embodiment has the same components as the remote support device 20 described in the first embodiment, and only the differences will be described with reference to FIG. 3 described above.

The automatic determination unit 21B is associated with each determination scene and the location associated with the determination scene, each time zone associated with the determination scene and the determination scene, or the determination scene and the determination scene. Calculate the duration distribution for each vehicle type.

14 (A) to 14 (C) are diagrams for explaining a method of calculating the duration distribution for each place associated with the determination scene according to the fourth embodiment.

As shown in FIG. 14A, the determination scene is further divided by a grid to calculate the duration distribution. In the example of FIG. 14A, the duration distribution and the effective period reflecting different characteristics (regional characteristics, etc.) in areas 1 to 4 are calculated. Therefore, the accuracy of the automatic response is improved.

As shown in FIG. 14B, when the judgment scene is "parking on the street", the duration distribution is calculated by dividing the judgment scene into facilities (convenience stores, commercial facilities, houses, parks, etc.) facing the judgment scene. .. In the example of FIG. 14B, the purpose of on-street parking differs for each facility, and the parking time also differs depending on the purpose. Therefore, the duration distribution and the valid period that reflect the characteristics of parking time that differ for each facility are calculated. Therefore, the accuracy of the automatic response is improved.

As shown in FIG. 14 (C), when the judgment scene is "accident", "obstacle", etc., the distance from the base that provides the road service such as JAF (JAPAN AUTOMOBILE FEDERATION). Divide by and calculate the duration distribution. In the example of FIG. 14C, the duration distribution and the effective period reflecting the characteristics of the time required for removal, which are different for each area, are calculated.

FIG. 15 is a diagram showing an example of a duration model for a distance from a facility.

In FIG. 15, the occurrence frequency (Y-axis), the duration (X-axis), and the distance from the base (Z-axis) are set as three axes, and the three-dimensional duration distribution is used for machine learning such as a Gaussian process. Estimate using. In this case, it is possible to estimate a certain distance from the information of another distance even if the information is missing.

FIG. 16 is a diagram for explaining a method of calculating a duration distribution for each time zone associated with the determination scene according to the fourth embodiment.

As shown in FIG. 16, for example, the determination scene of "parking on the street" is further divided into time zones (for example, noon: 8:00 to 20:00, night: 20:00 to 8:00), and the duration distribution is calculated. Similarly, the determination scene of the "obstacle" (or "breakdown vehicle") is further divided by the time zone (for example, noon: 8:00 to 20:00, night: 20:00 to 8:00), and the duration distribution is calculated. .. For example, the duration of "street parking" is short during the day and long at night. On the other hand, the duration of an "obstacle" (or "breakdown vehicle") is short during the day and long during the night because it is easy to handle wreckers such as removal. In the example of FIG. 16, the duration distribution and the effective period reflecting different characteristics for each time zone are calculated. Therefore, the accuracy of the automatic response is improved.

In addition, in the "construction" determination scene, the duration distribution is calculated by dividing by the amount of precipitation with respect to the duration from registration. The duration distribution and effective period that reflect different characteristics for each period, such as precipitation, are calculated. Therefore, the accuracy of the automatic response is improved.

FIG. 17 is a diagram showing an example of a duration model for precipitation.

In FIG. 17, the occurrence frequency (Y-axis), duration (X-axis), and precipitation (Z-axis) are set as three axes, and the duration distribution in three dimensions is obtained by using machine learning such as a Gaussian process. presume. In this case, it is possible to estimate a certain amount of precipitation from the information of another amount of precipitation even if the information is missing.

FIG. 18 is a diagram for explaining a method of calculating the duration distribution for each vehicle type associated with the determination scene according to the fourth embodiment.

As shown in FIG. 18, in the "street parking" determination scene, the duration distribution is calculated by further dividing by vehicle type (for example, passenger car, taxi, truck, etc.). For example, in the case of a truck, the parking time on the street is long. In the example of FIG. 18, the duration distribution and the effective period reflecting the characteristics different for each vehicle type are calculated. Therefore, the accuracy of the automatic response is improved.

As described above, according to the present embodiment, the duration distribution is calculated for each place, each time zone, or each vehicle type associated with the determination scene. Therefore, the accuracy of the automatic response is improved.

[Fifth Embodiment]
FIG. 19 is a block diagram showing an example of a functional configuration of the vehicle-mounted device 10 and the remote support device 20A according to the fifth embodiment.

As shown in FIG. 19, the CPU 11 of the vehicle-mounted device 10 according to the present embodiment functions as a request output unit 11A and a determination acquisition unit 11B.

On the other hand, the CPU 21 of the remote support device 20A according to the present embodiment functions as a request acquisition unit 21A, an automatic determination unit 21B, a determination output unit 21C, an operator presentation unit 21D, a duration calculation unit 21E, and an operator determination registration unit 21F. .. The remote support device 20A according to the present embodiment is different from the remote support device 20 shown in FIG. 3 above in that it does not have the duration distribution DB 25C.

As shown in FIG. 19, the remote support device 20A according to the present embodiment calculates the duration distribution and the valid period of the judgment scene by the past operator in advance, and the calculated duration distribution and the valid period are the operator judgment log DB 25B. To register with. Then, when the valid period is set, the valid period is read from the operator determination log DB 25B and set.

That is, the operator determination registration unit 21F operates the duration calculation unit 21E when the same determination scene is not registered in the same area as the area for which the support request was requested this time. Then, the duration calculation unit 21E calculates the duration length and the duration distribution for the determination scene by the operator who is going to be registered by the operator determination registration unit 21F. Then, the duration calculation unit 21E associates the calculated duration distribution with the operator determination log and registers it in the operator determination log DB 25B. At this time, the validity period may be registered instead of the duration distribution, or the duration distribution and the validity period may be registered.

The remote support device according to the embodiment has been illustrated and described above. The embodiment may be in the form of a program for causing a computer to execute the functions of each part included in the remote support device. The embodiment may be in the form of a non-transitional recording medium that can be read by a computer that stores these programs.

In addition, the configuration of the remote support device described in the above embodiment is an example, and may be changed depending on the situation within a range that does not deviate from the gist.

Further, the processing flow of the program described in the above embodiment is also an example, and even if unnecessary steps are deleted, new steps are added, or the processing order is changed within a range that does not deviate from the purpose. good.

Further, in the above embodiment, the case where the processing according to the embodiment is realized by the software configuration by using the computer by executing the program has been described, but the present invention is not limited to this. The embodiment may be realized, for example, by a hardware configuration or a combination of a hardware configuration and a software configuration.

10 In-vehicle device, 11, 21 CPU, 11A request output unit, 11B judgment acquisition unit, 12, 22 memory, 13, 24 display unit, 14, 25 storage unit, 15 sensor group, 16 camera, 17, 26 communication unit, 20 , 20A remote support device, 23 operation unit, 21A request acquisition unit, 21B automatic judgment unit, 21C judgment output unit, 21D operator presentation unit, 21E duration calculation unit, 21F operator judgment registration unit, 25A remote support program, 25B operator judgment Log DB, 25C duration distribution DB, 25D scene classification table, 100 remote support system

Claims (13)

A remote support device (20) that provides remote support to autonomous vehicles in the area managed by the autonomous driving support center.
The acquisition unit (21A) that acquires support request information including the position information of the autonomous driving vehicle from the autonomous driving vehicle, and
From the database (25B) including the operator judgment log, which is the log of the judgment scene that the operator has responded to in the past, in the specific area including the point indicated by the position information according to the support request information acquired by the acquisition unit. A classification unit (21B) that acquires a determination scene most recently responded by the operator and classifies the scene of the automatically driving vehicle into either a scene to be overtaken or a scene to wait based on the acquired determination scene. When,
When the scene of the autonomous driving vehicle is classified into the scene to be overtaken by the classification unit, the waiting period represents a period for outputting an automatic response based on the operator's latest response as a valid response. The setting unit (21B) that sets the setting shorter than when it is classified into the scene to be
Remote support device equipped with. When the period from the latest response by the operator to the acquisition of the support request information is within the valid period, an instruction is given to output the automatic response to the autonomous driving vehicle, and the support is provided from the latest response by the operator. The first aspect of claim 1 is further provided with an instruction unit (21B) for stopping the automatic response and instructing the operator to register the determination scene this time when the period until the request information is acquired exceeds the valid period. Remote support device. In the setting unit, when the specific area is a region where the frequency of occurrence of the scene to be waited is equal to or higher than a preset threshold value, or the frequency of occurrence of the scene to be waited is the occurrence of the scene to be overtaken. The remote support device according to claim 1 or 2, wherein when the area is higher than the frequency, the valid period of the scene to be overtaken is set shorter than the valid period of the scene to be waited. In the setting unit, when the specific area is a region where the frequency of occurrence of the scene to be overtaken is equal to or higher than a preset threshold value, or the frequency of occurrence of the scene to be overtaken is the occurrence of the scene to be waited for. The remote support device according to claim 1 or 2, wherein when the area is higher than the frequency, the valid period of the scene to be overtaken is set to be equal to or longer than the valid period of the scene to be waited for. The operator determination log further includes first sensing information obtained from the most recent autonomous vehicle that has received the operator's most recent response.
The support request information further includes the second sensing information obtained from the autonomous driving vehicle.
The classification unit classifies the scene of the autonomous driving vehicle into either the scene to be overtaken or the scene to be on standby based on the degree of similarity between the first sensing information and the second sensing information. The remote support device according to claim 1. When the remaining time of the valid period with respect to the period from the latest response by the operator to the acquisition of the support request information is a certain time or more and the similarity is equal to or more than the first threshold value, or by the operator. When the remaining time of the valid period with respect to the period from the latest response to the acquisition of the support request information is less than the fixed time, and the similarity is greater than or equal to the second threshold value larger than the first threshold value. The remote support device according to claim 5, further comprising an instruction unit that gives an instruction to output the automatic response to the automatic driving vehicle. When the similarity is equal to or higher than the threshold value, if the period from the latest response by the operator to the acquisition of the support request information is within the valid period, an instruction is given to output the automatic response to the autonomous driving vehicle. When the similarity is less than the threshold value, the automatic response is stopped even if the period from the latest response by the operator to the acquisition of the support request information is within the valid period, and the operator is determined this time. The remote support device according to claim 5, further comprising an instruction unit for instructing the scene to be registered. The remote support device according to claim 7, wherein the threshold value becomes larger as the remaining time of the valid period with respect to the period from the latest response by the operator to the acquisition of the support request information decreases. Using the operator judgment log, a calculation unit (21E) for calculating a duration distribution representing the relationship between the occurrence frequency of the judgment scene and the duration of the judgment scene is further provided for each judgment scene.
The remote according to any one of claims 1 to 8, wherein the setting unit sets the validity period according to the scene of the autonomous driving vehicle based on the duration distribution calculated by the calculation unit. Support device. The calculation unit is used for each determination scene and a location associated with the determination scene, or for each determination scene and a time zone associated with the determination scene, or a vehicle type associated with the determination scene and the determination scene. The remote support device according to claim 9, which calculates the duration distribution for each. The remote support according to claim 9 or 10, wherein the calculation unit calculates the validity period based on the duration distribution and registers the calculated validity period in the database in association with the operator determination log. Device. It is a remote support method using a remote support device that provides remote support to autonomous vehicles in the area managed by the autonomous driving support center.
From the autonomous driving vehicle, support request information including the position information of the autonomous driving vehicle is acquired, and the support request information is obtained.
In response to the acquired support request information, the operator most recently responds in a specific area including the point indicated by the location information from the database including the operator determination log, which is the log of the determination scene that the operator has responded to in the past. The determined judgment scene is acquired, and based on the acquired determination scene, the scene of the autonomous driving vehicle is classified into either a scene to be overtaken or a scene to be waited for.
When the scene of the autonomous driving vehicle is classified into the scene to be overtaken, the valid period representing the period for outputting the automatic response based on the operator's latest response as a valid response is classified into the scene to be waited. Set shorter than if
Remote support method. It is a remote support program (25A) of a remote support device that provides remote support to autonomous vehicles in the area managed by the autonomous driving support center.
Computer,
The acquisition unit (21A), which acquires support request information including the position information of the autonomous driving vehicle from the autonomous driving vehicle,
From the database (25B) including the operator judgment log, which is the log of the judgment scene that the operator has responded to in the past, in the specific area including the point indicated by the position information according to the support request information acquired by the acquisition unit. A classification unit (21B) that acquires a determination scene most recently responded by the operator and classifies the scene of the automatically driving vehicle into either a scene to be overtaken or a scene to wait based on the acquired determination scene. ,as well as,
When the scene of the autonomous driving vehicle is classified into the scene to be overtaken by the classification unit, the waiting period represents a period for outputting an automatic response based on the operator's latest response as a valid response. Setting unit (21B), which is set shorter than when it is classified into a power scene,
Remote support program to function as.

Images