control device

Abstract

We provide technology that enables stable autonomous driving control. [Solution] The control device (100) includes an event detection unit (110) and a control rule modification unit (120). The event detection unit (110) determines whether an event that triggers a change in the control rule during autonomous driving of the vehicle has been detected while the vehicle is autonomously driving using a first control rule based on machine learning. If the event detection unit (110) detects a trigger event, the control rule modification unit (120) changes the control rule during autonomous driving of the vehicle to a second control rule corresponding to the trigger event.

Description

Technical Field

[0001] The present invention relates to a control device, a control method, and a program.

Background Art

[0002] In recent years, research and development on automatic driving control have been advanced. For example, in Patent Document 1 below, a technique for performing risk prediction and the like based on a knowledge base storing logical expressions generated using a known supervised machine learning method and applying it to automatic driving control of a vehicle is disclosed.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] A model (machine learning model) constructed by machine learning changes according to input learning data and may have unique characteristics for each machine learning model. Such a machine learning model can basically be said to be preferable in that optimal control based on the learning data given when constructing the model can be executed. On the other hand, appropriate countermeasures are not always executed for input data (events) that deviate significantly from the learning data given when constructing the model. From this point, in a scene where stable control is required, control based on a machine learning model may not be preferable.

[0005] As an example of the problem to be solved by the present invention, there is provided a technique that enables stable automatic driving control.

Means for Solving the Problems

[0006] The invention described in claim 1 is, An event detection unit determines whether an event that triggers a change in the control rule during autonomous driving of the vehicle has been detected while the vehicle is autonomously driving using a first control rule based on machine learning, When the event detection unit detects the trigger event, the control rule change unit changes the control rule for the vehicle's autonomous driving to a second control rule corresponding to the trigger event. This is a control device equipped with [a specific feature / feature].

[0007] This disclosure includes, A control method performed by a computer, The process involves determining whether an event that triggers a change in the control rules for the vehicle's autonomous driving is detected while the vehicle is autonomously driving using a first control rule based on machine learning, and When the event detection unit detects the trigger event, the control rule for the vehicle's autonomous driving is changed to a second control rule corresponding to the trigger event. This includes control methods that include [specific details].

[0008] This disclosure includes, Computers, An event detection means for determining whether an event that triggers a change in the control rules during autonomous driving of a vehicle has been detected while the vehicle is autonomously driving using a first control rule based on machine learning, and When the event detection unit detects the trigger event, a control rule changing means changes the control rule for the vehicle's automatic driving to a second control rule corresponding to the trigger event. It includes a program to make it function as such. [Brief explanation of the drawing]

[0009] The aforementioned objectives, as well as other objectives, features, and advantages, will become even clearer from the preferred embodiments described below and the accompanying drawings.

[0010] [Figure 1] This is a diagram illustrating the general outline of the control device according to the present invention. [Figure 2] This is a block diagram conceptually showing the functional configuration of the control device in the first embodiment. [Figure 3] This figure illustrates the hardware configuration of the control device according to the first embodiment. [Figure 4] This is a flowchart illustrating the processing flow performed by the control device of the first embodiment. [Figure 5] This diagram illustrates information that associates a given event with a second control rule. [Figure 6] This is a block diagram conceptually showing the functional configuration of the control device in the second embodiment. [Figure 7] This figure illustrates the hardware configuration of the control device according to the second embodiment. [Figure 8] This is a sequence diagram illustrating the processing flow executed by the control device of the second embodiment. [Modes for carrying out the invention]

[0011] [Overview] FIG. 1 is a diagram for explaining an overview of a control device 100 according to the present invention. In the example of FIG. 1, the control device 100 is a device (e.g., an ECU (Electronic Control Unit), etc.) mounted on a vehicle V. The control device 100 can change a first control rule that changes (optimizes) by machine learning and a second control rule that is a fixed rule independent of machine learning. As will be described in detail later, when the automatic driving of the vehicle is performed using the first control rule based on machine learning, the control device 100 detects a predetermined event that triggers a change in the control rule of the automatic driving. If so, it changes to a second control rule, which is a fixed rule corresponding to the event. Note that information regarding a predetermined event that triggers a change in the control rule of the automatic driving can be obtained from the output of a sensor device 300 mounted on the vehicle V or from an external device 500. The external device 500 is, for example, a device similar to the control device 100 mounted on another vehicle not shown, a device for vehicle-to-road communication provided along the road, or the like.

[0012] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In all the drawings, the same components are denoted by the same reference numerals, and the description will be omitted as appropriate. Also, unless otherwise specified, each block in the block diagram represents a configuration in terms of functional units, not hardware units.

[0013] [First Embodiment] FIG. 2 is a block diagram conceptually showing the functional configuration of the control device 100 in the first embodiment. As shown in FIG. 2, the control device 100 of the present embodiment includes an event detection unit 110 and a control rule change unit 120.

[0014] The event detection unit 110 determines whether a predetermined event has been detected while the vehicle is operating autonomously using a first control rule based on machine learning. The predetermined event is an event that triggers a change in the control rule during autonomous driving of the vehicle. The predetermined event can also be described as an event that indicates the timing to interrupt the control of autonomous driving using the first control rule based on machine learning. Specific examples of predetermined events will be described later. The event detection unit 110 can detect a predetermined event based on the results of analyzing the output from various sensor devices 300 mounted on the vehicle. The event detection unit 110 may also detect a predetermined event via an external device (a control device mounted on another vehicle in the vicinity, or a vehicle-to-infrastructure communication device installed along the road) by communicating with that external device. The event detection unit 110 may also detect a predetermined event based on dynamic information contained in the autonomous driving map data used during autonomous driving of the vehicle V. Here, the autonomous driving map data is, for example, map data called a "dynamic map," and is data that includes conventional map information (static information) and information that changes in real time (dynamic information). Dynamic information includes information that can change over relatively short spans (e.g., seconds), such as ITS (Intelligent Transport Systems) lookup information (e.g., surrounding vehicles, pedestrian information, signal information), and information that can change over somewhat shorter spans (e.g., minutes), such as accident information, traffic congestion information, and local weather information (this is also called "quasi-dynamic information"). Static information includes information that can change over relatively long spans (e.g., months), such as road surface information, lane information, and 3D structures, and information that can change over somewhat longer spans (e.g., hours), such as traffic regulation information, road construction information, and wide-area weather information (this is also called "quasi-static information"). In this invention, quasi-static information may be classified under the category of dynamic information. Note that the static and dynamic information included in the map data for autonomous driving is not limited to the examples given here. Map data for autonomous driving may include various types of information that can be used for autonomous driving control of vehicles.

[0015] While the vehicle is under autonomous driving using the first control rule based on machine learning, when an event is detected by the event detection unit 110, the control rule for the autonomous driving of the vehicle is changed to a second control rule, which is a fixed rule corresponding to the detected event.

[0016] The first control rule used by each vehicle during autonomous driving is a rule based on machine learning as described above. Therefore, the behavior of the vehicle when using the first control rule may exhibit unique characteristics according to the learning results of the given training data. When the control during autonomous driving is performed using the first control rule, there is a possibility that unpredictable and unstable behavior may occur due to the unique characteristics of this first control rule. In this regard, in the present embodiment, when a predetermined event is detected while autonomous driving is being performed using the first control rule based on machine learning, the control rule for the autonomous driving is changed to a second control rule, which is a fixed rule corresponding to the detected event. As a result, when a predetermined event is detected, the behavior of the vehicle during autonomous driving is controlled according to the fixed rule, so that unpredictable and unstable behavior can be suppressed.

[0017] Furthermore, in locations where multiple vehicles equipped with the control device 100 are gathered, each vehicle can perform autonomous driving using a fixed second control rule instead of a first control rule based on machine learning results. This allows for control over the movement of each vehicle, and as a result, is expected to improve the traffic environment. For example, consider a three-lane road where there is an obstacle in one lane, and each vehicle traveling in that lane needs to change lanes to avoid the obstacle. In this case, if each vehicle were to act according to a first control rule with characteristics unique to each vehicle, problems such as congestion occurring or worsening due to vehicles concentrating in one of the remaining two lanes could arise. In such cases, the control device 100 of this embodiment can control each vehicle by switching the autonomous driving control rule of each vehicle to a fixed rule corresponding to the detected event (in this case, for example, a rule such as "as an obstacle avoidance action, move to a different lane than the one the vehicle in front has moved to"), thereby distributing vehicles to each of the remaining two lanes and minimizing congestion.

[0018] The control device 100 of this embodiment will be described in more detail below.

[0019] [Hardware configuration] Each functional component of the control device 100 may be implemented by hardware (e.g., hardwired electronic circuits) or by a combination of hardware and software (e.g., a combination of an electronic circuit and a program to control it). The following will further explain the case where each functional component of the control device 100 is implemented by a combination of hardware and software.

[0020] Figure 3 is a diagram illustrating the hardware configuration of the control device 100 according to the first embodiment. Computer 200 is a computer that implements the control device 100. For example, computer 200 is an ECU (Electronic Control Unit) capable of controlling the operation of a vehicle during autonomous driving. Computer 200 may be a computer specifically designed to implement the control device 100, or it may be a general-purpose computer.

[0021] Computer 200 includes a bus 202, a processor 204, a memory 206, a storage device 208, an input / output interface 210, and a network interface 212. The bus 202 is a data transmission path for the processor 204, memory 206, storage device 208, input / output interface 210, and network interface 212 to send and receive data to and from each other. However, the method of connecting the processor 204 and the other components is not limited to bus connection. The processor 204 is an arithmetic processing unit implemented using a microprocessor or the like. The memory 206 is a main memory device implemented using RAM (Random Access Memory) or the like. The storage device 208 is an auxiliary memory device implemented using ROM (Read Only Memory) or flash memory or the like.

[0022] The input / output interface 210 is an interface for connecting the computer 200 to peripheral devices. Various analog and digital signals used for vehicle control are input to or output from the computer 200 via the input / output interface 210. The input / output interface 210 may include, as appropriate, an A / D converter for converting analog input signals to digital signals and a D / A converter for converting digital output signals to analog signals.

[0023] For example, in Figure 3, the input / output interface 210 is connected to sensor devices 300 and drive circuits 400 used for vehicle control. Sensor devices 300 include LIDAR (Light Detection and Ranging), millimeter-wave radar, sonar, and cameras. Although not shown, multiple sensor devices 300 can be connected to the computer 200 via the input / output interface 210. The drive circuit 400 is a circuit for driving various mechanisms of the vehicle, such as the gears, engine, and steering. The control device 100 can control the operation of the vehicle during autonomous driving by controlling the operation of the drive circuit 400.

[0024] The network interface 212 is an interface for connecting the computer 200 to a communication network. This communication network may be, for example, a CAN (Controller Area Network), a LAN (Local Area Network), or a WAN (Wide Area Network). The network interface 212 may connect to the communication network via wireless or wired connection. The computer 200 can communicate with the control devices 502 and vehicle-to-infrastructure communication devices 504 of other vehicles via wireless LAN or the like, and can obtain information about events used in the processing of the control device 100 from these devices.

[0025] The storage device 208 stores program modules for realizing each functional component of the control unit 100. The processor 204 reads these program modules into memory 206 and executes them to realize the functions of the control unit 100. The storage device 208 may also store map data for autonomous driving used when the vehicle V is in autonomous driving mode.

[0026] [Processing flow] The processing flow performed by the control device 100 of this embodiment will be schematically explained using Figure 4. Figure 4 is a flowchart illustrating the processing flow performed by the control device 100 of the first embodiment.

[0027] First, when the vehicle's driving mode switches to automatic driving mode (S102:YES), the event detection unit 110 is activated and begins monitoring for predetermined events (S104). Subsequently, if the event detection unit 110 detects a predetermined event (S104:YES), the event detection unit 110 notifies the control rule change unit 120 that a predetermined event has been detected (S106).

[0028] The control rule modification unit 120 identifies a second control rule corresponding to the event notified in the processing of S106 (S108). The control rule modification unit 120 can identify a second control rule corresponding to an event detected by the event detection unit 110 using, for example, a table as shown in Figure 5. The table illustrated in Figure 5 stores the identification information of an event and the identification information of the second control rule to be applied in response to the detection of that event in association with each other. For example, the control rule modification unit 120 can obtain the identifier of the event detected in the processing of S106 from the event detection unit 110, and then identify the second control rule by referring to the table in Figure 5 based on that event identifier. The control rule modification unit 120 then forwards an instruction to apply the second control rule read in the processing of S108 to an ECU or other device that controls autonomous driving (S110). As a result, the operation of the vehicle during autonomous driving is controlled based on the second control rule. Note that this explanation is merely illustrative, and the operation of the control rule modification unit 120 is not limited to using the table illustrated in Figure 5. For example, Figure 5 shows an example where a different second control rule is associated with each event, but this is not the only example; the same second rule may be associated with multiple events.

[0029] Below, we will provide some specific examples to explain the operation in more detail.

[0030] <First specific example> This specific example describes a case in which the event detection unit 110 detects "an abnormality has occurred in the sensor device 300" as a predetermined event. Here, an abnormality in the sensor device 300 refers to an abnormality in the output signal from the sensor device 300 or a communication failure between the sensor device 300 and the control device 100, due to contamination of the optical system (lens, etc.), internal failure, disturbances in the sensing environment (sunlight, rain, fog, snow, headlights of oncoming vehicles, etc.), or the detection of an error signal or an unexpected signal.

[0031] Next, we will describe a case in which the event detection unit 110 detects a predetermined event based on "dynamic information included in the autonomous driving map data." As mentioned above, the autonomous driving map data is a digital map that incorporates not only static information (map information such as road surface information, lane information, and 3D structures) but also dynamic information (accident information, traffic congestion information, weather information, pedestrian information, traffic signal information, etc.). Specifically, the dynamic information included in the autonomous driving map data refers to the dynamic information described above. This dynamic information is distributed to the vehicle V from a server that manages accident information, etc. The vehicle V, upon receiving accident information, etc., stores the accident information, etc., in an area of ​​the autonomous driving map data that indicates dynamic information. The event detection unit 110 can then detect an event by referring to the dynamic information included in the autonomous driving map data used when the vehicle V is autonomously driving.

[0032] The event detection unit 110 monitors the signal lines connected to the sensor device 300 via the input / output interface 210 and measures the intensity of the signal output from the sensor device 300 and analyzes the content of the signal. The event detection unit 110 then determines whether the measured signal intensity is lower than a predetermined reference value, or whether the analyzed signal is an error signal or an unexpected signal. In this case, the predetermined reference value to be compared is pre-stored in, for example, memory 206 or storage device 208. If it is detected that the measured signal intensity is below the predetermined reference value, or that the analyzed signal is an error signal or an unexpected signal, the event detection unit 110 notifies the control rule modification unit 120 that "an abnormality has occurred in the sensor device 300." Based on the notification from the event detection unit 110, the control rule modification unit 120 identifies a second control rule to be applied when "an abnormality has occurred in the sensor device 300." The second control rule in this case is not particularly limited, but could be, for example, a rule such as "stop the vehicle using a predetermined procedure (for example, turning on the hazard lights and controlling the brakes to gradually decelerate)."

[0033] If the output signal strength of the sensor device 300 decreases due to a malfunction, the probability of the autonomous driving control processing unit misinterpreting the surrounding conditions of the vehicle increases, making autonomous driving operation more likely to become unstable. When a malfunction of the sensor device 300 is detected, autonomous driving control can be performed according to a second control rule in which a fixed control operation is defined, thereby preventing unstable operation during autonomous driving. Alternatively, when a malfunction of the sensor device 300 is detected, the control device 100 may be configured to abandon autonomous driving control and delegate control authority to the driver instead of changing to the second control rule.

[0034] <Second specific example> This specific example describes a case in which the event detection unit 110 detects as a predetermined event that "an accident has occurred in at least one of the lanes when a vehicle is traveling on a road with multiple lanes on one side."

[0035] The event detection unit 110 can detect a predetermined event by acquiring information indicating the location of the accident vehicle (e.g., location coordinates on a map, information about the lane where the accident occurred, etc.) via, for example, the control device 502 of another vehicle or a vehicle-to-infrastructure communication device 504 installed along the road. The control device 502 of another vehicle, for example, when it detects the presence of the accident vehicle using various sensors mounted on that vehicle, generates information indicating the presence of the accident vehicle along with the location information of the accident vehicle (e.g., location coordinates on a map, information about the lane where the accident occurred, etc.), and can transmit this information via vehicle-to-vehicle communication. In this case, the event detection unit 110 can detect a predetermined event via the control device 502 of the other vehicle by performing vehicle-to-vehicle communication with the control device 502 of the other vehicle. Furthermore, the vehicle-to-infrastructure communication device 504 can collect information indicating the presence of the accident vehicle and the location information of the accident vehicle from the control device 502 of the other vehicle, and broadcast the collected information within the area under the jurisdiction of the vehicle-to-infrastructure communication device 504. In this case, the event detection unit 110 can detect a predetermined event via the vehicle-to-infrastructure communication device 504 by receiving information broadcast from the vehicle-to-infrastructure communication device 504. Furthermore, if the sensor device 300 mounted on the vehicle is a camera with an image sensor, the presence or absence of an accident vehicle (a predetermined event) can be detected by analyzing the image data generated by the camera. Similarly, if the sensor device 300 mounted on the vehicle is a LiDAR, the presence or absence of an accident vehicle (a predetermined event) can be detected based on an image generated from point cloud data obtained by laser scanning of the LiDAR. For example, the event detection unit 110 can determine the presence or absence of an accident vehicle in the image data by using a CNN (Convolutional Neural Network) constructed using images of accident vehicles as training data.

[0036] When the event detection unit 110 detects the predetermined event, the control rule change unit 120, for example, changes the control rule used during autonomous driving to the default rule that was in place before the first control rule was constructed by machine learning. Here, the default rule is the control rule in its initial state before machine learning has been performed; in other words, it is a control rule that does not have any unique characteristics. By controlling autonomous driving using such a default rule, it is possible to suppress unpredictable and unstable behavior caused by unique characteristics that arise from machine learning.

[0037] As another example, if the purpose is to control the operation of multiple vehicles, when the event detection unit 110 detects the predetermined event, the control rule change unit 120 may change the control rule used during autonomous driving to a rule that is commonly used among multiple control devices (control device 100, control devices 502 of other vehicles, and other control devices installed in vehicles not shown) (hereinafter also referred to as "common rule"). Common rules can be prepared as rules common to the whole world, by country, by region, by vehicle type, or by vehicle manufacturer. These common rules are pre-stored in memory 206 or storage device 208 in a format such as that shown in Figure 5. Alternatively, common rules may be stored in autonomous driving map data stored in storage device 208 or an external server device of vehicle V. By using common rules among multiple control devices, variations in the operation of each vehicle during autonomous driving are reduced, and the operation of each vehicle can be controlled. Specific examples of common rules are not limited to, but include, "move to a different lane than the one the vehicle in front has moved to."

[0038] <Third specific example> This specific example describes a case in which the event detection unit 110 detects "the presence of an obstacle (such as a fallen object, a pothole or flooding in the road surface) in the lane where the vehicle is traveling" as a predetermined event.

[0039] The event detection unit 110 can detect a predetermined event by acquiring information indicating the location of an obstacle (e.g., location coordinates on a map, information about the lane in which the obstacle is located) via, for example, the control device 502 of another vehicle or a vehicle-to-infrastructure communication device 504 installed along the road. The control device 502 of another vehicle, for example, when it detects an obstacle using various sensors mounted on that vehicle, generates information indicating the presence of the obstacle, along with the location information of the obstacle (e.g., location coordinates on a map, information about the lane in which the obstacle is located), and can transmit this information via vehicle-to-vehicle communication. In this case, the event detection unit 110 can detect a predetermined event via the control device 502 of the other vehicle by performing vehicle-to-vehicle communication with the control device 502 of the other vehicle. Furthermore, the vehicle-to-infrastructure communication device 504 can collect information indicating the presence of an obstacle and the location information of the obstacle from the control device 502 of the other vehicle and broadcast the collected information within the area under the jurisdiction of the vehicle-to-infrastructure communication device 504. In this case, the event detection unit 110 can detect a predetermined event via the vehicle-to-infrastructure communication device 504 by receiving information broadcast from the vehicle-to-infrastructure communication device 504. It can also detect the presence or absence of obstacles (a predetermined event) using the sensor device 300 mounted on the vehicle. For example, the event detection unit 110 can recognize the shape of the road surface and obstacles on the road surface based on images generated using an image sensor or scanning results (distance images) from a LiDAR.

[0040] In this case, the control rule modification unit 120 can change the control rules used during autonomous driving to default rules used before the first control rules were constructed by machine learning, or to rules commonly used among multiple control devices, similar to the second specific example.

[0041] <Fourth specific example> This specific example describes a case in which the event detection unit 110 detects as a predetermined event that "information indicating the actions that the vehicle should take has been received as a second control rule from the control device 502 or vehicle-to-infrastructure communication device 504 of another vehicle."

[0042] As an example, first, the control device 502 of the other vehicle detects the presence or absence of obstacles or accident vehicles on the road surface. The control device 502 of the other vehicle can detect obstacles or accident vehicles on the road surface based on, for example, the output from a sensor device installed in the other vehicle. The control device 502 of the other vehicle can also obtain information about obstacles on the road surface or accident vehicles from the vehicle-to-infrastructure communication device 504. When an obstacle or accident vehicle is detected on the road surface, the control device 502 of the other vehicle controls the other vehicle to avoid the obstacle or accident vehicle and generates information indicating the actions that the following vehicle should take. Then, the control device 502 of the other vehicle assigns, for example, a dedicated identifier to the generated information and transmits it to the following vehicle via the communication device. As another example, the vehicle-to-infrastructure communication device 504 may detect the presence or absence of obstacles or accident vehicles on the road surface based on the output from a sensor device installed in the other vehicle. The vehicle-to-infrastructure communication device 504 collects information (sensor information) from surrounding vehicles regarding obstacles on the road surface or accident vehicles, for example, and uses the collected information to determine the actions that each vehicle in the target area should take. The vehicle-to-infrastructure communication device 504 assigns a dedicated identifier to the information indicating the determined action and broadcasts it within the target area. As a specific example of instructions, if an obstacle is detected in the center lane on a three-lane road, the control device 502 of another vehicle or the vehicle-to-infrastructure communication device 504 can send an instruction that vehicles with odd-numbered license plate digits at the end should move to the left lane, and vehicles with even-numbered digits should move to the right lane. When the event detection unit 110 receives information indicating the action that its own vehicle should take from the control device 502 of another vehicle or the vehicle-to-infrastructure communication device 504, it uses this information as a second control rule to control the operation of its own vehicle during autonomous driving. Furthermore, the event detection unit 110 can determine that the information received from another vehicle is "information indicating the action that the vehicle should take" based on the identifier assigned to the information. In this specific example, the operation during autonomous driving is controlled based on instructions transmitted from the control device 502 and the vehicle-to-infrastructure communication device 504 of the other vehicle. This prevents unstable operation caused by the unique characteristics of machine learning.

[0043] [Second Embodiment] In the first embodiment, an example was shown in which a common rule is used among multiple control devices (control device 100, control device 502 of another vehicle, and other control devices mounted on vehicles not shown). The control device 100 of this embodiment further includes a configuration for updating this common rule.

[0044] [Functional Configuration] Figure 6 is a block diagram conceptually showing the functional configuration of the control device 100 in the second embodiment. As shown in Figure 6, the control device 100 in this embodiment further includes an update information acquisition unit 130 and a common rule update unit 140. Also, as shown in Figure 7, in this embodiment, a server device 506 that generates information for updating common rules is connected to the control device 100 in a communicative manner.

[0045] The update information acquisition unit 130 acquires update information for common rules from the server device 506. For example, the server device 506 learns from the results of automated driving control based on common rules (e.g., changes in congestion level) as rewards and updates the common rules. Specifically, the server device 506 acquires information from sensor devices installed in each vehicle and around the road that shows how each vehicle operated according to the common rules, as well as information that shows changes in congestion level, and evaluates the current common rules. Then, the server device 506 updates the common rules based on the evaluation results. For example, suppose that as a result of "action A" defined in the common rules being performed, a reward is obtained that "the congestion level has worsened beyond an acceptable value." In this case, information is generated to update the common rules so that a certain penalty, or a penalty corresponding to the degree of deterioration, is given to "action A," thereby lowering the priority of selecting action A. In addition to the example given here, it is also possible to adopt a configuration in which a manager manually inputs update information for common rules into the server device 506 and distributes it to the control devices of each vehicle. The server device 506 distributes the updated common rule information generated in this way to each vehicle. The server device 506 may be configured to distribute the updated common rules, which have been updated using the common rule update information, to the control devices of each vehicle. Furthermore, the common rule update information distributed from the server device 506 may be distributed to the control devices 100 of each vehicle via the vehicle-to-infrastructure communication device 504.

[0046] The common rule update unit 140 updates the common rules stored in the memory 206 and storage device 208 using the update information of the common rules acquired by the update information acquisition unit 130.

[0047] [Hardware configuration] Figure 7 illustrates the hardware configuration of the control device 100 in the second embodiment. In this embodiment, the storage device 208 stores program modules for realizing the functions of the update information acquisition unit 130 and the common rule update unit 140. The processor 204 reads these program modules into the memory 206 and executes them to realize the functions of the update information acquisition unit 130 and the common rule update unit 140, respectively. In this embodiment, a server device 506 is connected via the network interface 212.

[0048] [Example of operation] The processing flow executed by the control device 100 of the second embodiment will be explained using Figure 8. Figure 8 is a sequence diagram illustrating the processing flow executed by the control device 100 of the second embodiment.

[0049] The server device 506 collects log information of automated driving control based on common rules from multiple control devices installed in each vehicle (for example, information indicating the actions and times on the common rules selected during automated driving) (S202). The server device 506 also acquires information indicating changes in congestion levels along with time information from the vehicle-to-infrastructure communication device 504 installed along the road (S204). Then, the server device 506 generates updated common rule information based on the actions selected by each vehicle according to the current common rules and the resulting results (rewards) (S206). The server device 506 can identify the correspondence between the information collected in the processes of S202 and S204 based on the time information. Then, the server device 506 distributes the updated common rule information generated in the process of S206 to the control devices of each vehicle (S208). The server device 506 may first send the updated common rule information to the vehicle-to-infrastructure communication device 504 and then distribute it to each vehicle via the vehicle-to-infrastructure communication device 504.

[0050] In each vehicle, the update information acquisition unit 130 receives the update information for the common rules distributed in S208. The common rule update unit 140 updates the current common rules stored in the memory 206 and storage device 208 based on the update information for the common rules acquired by the update information acquisition unit 130 (S210).

[0051] As described above, according to this embodiment, it is possible to optimize the common rules stored in each vehicle by receiving update information on common rules distributed from the server device 506.

[0052] The embodiments and examples described above with reference to the drawings are illustrative examples of the present invention, and various other configurations can also be adopted.

[0053] This application claims priority based on Japanese Patent Application No. 2017-138957, filed on 18 July 2017, and incorporates all of its disclosures herein.

Claims

[Claim 1] An event detection unit determines whether or not an event has been detected that triggers a change in the control rules for the vehicle's autonomous driving while the vehicle is autonomously driving using a first control rule based on machine learning. When the event detection unit detects the trigger event, the control rule change unit changes the control rule for the vehicle's automatic driving to a second control rule corresponding to the trigger event. A control device equipped with the following features.

Images