Automatic driving device, automatic driving method, and control program

Abstract

To provide a vehicle control device, a vehicle control method, an automatic driving device, and an automatic driving method capable of reducing danger that may occur during vehicle driving in vehicle control.SOLUTION: A vehicle control device is used in a vehicle capable of automatic driving and comprises: an environment information acquisition unit (23) for acquiring environment information relating to an environment in which an own vehicle is placed, other than obstacles around the own vehicle; a safe distance setting unit (281) for setting a minimum safe distance between the own vehicle and the obstacle to be secured in order to avoid contact between the own vehicle and the obstacle by using at least movement information of the own vehicle; and a safe distance adjusting unit (282) for increasing the safety distance that is set by the safe distance setting unit (281), in response to a latent occurrence probability which is a latent probability of potential occurrence of contact between the own vehicle and the obstacle based on the environment information acquired by the environment information acquisition unit (23) becomes higher.SELECTED DRAWING: Figure 1

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application is based on Patent Application No. 2019-186270 filed in Japan on October 9, 2019, and the contents of the original application are incorporated by reference in their entirety. [Technical field]

[0002] This disclosure , own Automatic driving device 、 Autonomous driving method and control program This is regarding. [Background technology]

[0003] Conventionally, automated driving, which automates the driving operation of a vehicle, is known. It is also considered to use a mathematical formula model that mathematically represents the concept of safe driving for automated driving.

[0004] For example, Patent Document 1 describes that in automated driving, a standard distance for evaluating safety (hereinafter, referred to as a safety distance) calculated by a mathematical formula model called the Responsibility Sensitive Safety (RSS) model is maintained at a minimum between the vehicle and obstacles such as other vehicles and pedestrians. Patent Document 1 also describes that the safety distance is determined from information related to the vehicle's behavior, such as speed, maximum deceleration, and maximum acceleration. The contents of the prior art documents are incorporated by reference as explanations of technical elements in this specification. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] International Publication No. 2018 / 115963 Summary of the Invention

[0006] In the technology disclosed in Patent Document 1, the safe distance is determined from information on the vehicle's behavior, but the possibility of approaching an obstacle varies depending on the surrounding environment other than the vehicle's behavior, such as the presence of places where pedestrians are likely to jump out. Therefore, the technology disclosed in Patent Document 1 is desired to be improved from the perspective of reducing the danger that may occur while the vehicle is traveling, such as by making it easier to avoid approaching obstacles.

[0007] One object of the present disclosure is to provide a vehicle control system capable of reducing dangers that may occur during vehicle operation. Self Automatic driving device 、 Autonomous driving method and control program The purpose of this invention is to provide

[0008] The above object is achieved by the combination of features recited in the independent claims, and the subclaims define further advantageous embodiments of the disclosure. The reference characters in parentheses in the claims indicate a correspondence with specific means described in the following embodiment as one aspect, and do not limit the technical scope of the present disclosure.

[0013] The autonomous driving device disclosed herein is used in an autonomous vehicle, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle, a safety distance setting unit that uses at least vehicle behavior information to set a minimum safety distance that the vehicle should keep between itself and an obstacle in order to avoid the vehicle coming into close contact with the obstacle, and a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as a potential occurrence probability, which is a potential occurrence probability of the vehicle coming into close contact with the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases. The safety distance setting unit sets at least a safety distance ahead of the vehicle, and the safety distance adjustment unit increases the safety distance ahead of the vehicle set by the safety distance setting unit as the potential occurrence probability increases. The safety distance setting unit also estimates a safety distance ahead for a vehicle following the vehicle as an obstacle, and when a following vehicle distance, which is a distance between the vehicle and the vehicle following the vehicle as an obstacle, is less than the safety distance ahead of the following vehicle estimated by the safety distance setting unit, the safety distance adjustment unit adds the distance by which the following vehicle distance falls short of the safety distance to the safety distance ahead of the vehicle set by the safety distance setting unit. do. Another autonomous driving device disclosed herein is used in a vehicle capable of autonomous driving, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, a safety distance setting unit that sets a minimum safety distance from the vehicle that the vehicle should maintain between itself and an obstacle using at least information on the vehicle's behavior to avoid the vehicle coming into close proximity with the obstacle, and a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as the potential occurrence probability, which is the potential probability of the vehicle coming into close proximity with the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases, and the safety distance setting unit sets at least the safety distances on the left and right sides of the vehicle, and includes a counting unit that counts the number of moving bodies detected around the vehicle, distinguishing between moving bodies on the left side and moving bodies on the right side of the vehicle, and when the potential occurrence probability is equal to or greater than a threshold, the safety distance adjustment unit increases only the safety distance on the side of the vehicle where the number of moving bodies counted by the counting unit is greater, out of the safety distances on the left and right sides of the vehicle set by the safety distance setting unit. Another autonomous driving device disclosed herein is used in a vehicle capable of autonomous driving, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, a safety distance setting unit that sets a minimum safety distance from the vehicle that the vehicle should maintain between itself and an obstacle using at least information on the behavior of the vehicle to avoid the vehicle coming into close proximity with the obstacle, a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as the potential occurrence probability, which is the probability of potential close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases, and a probability calculation unit that calculates the potential occurrence probability using the environmental information acquired by the environmental information acquisition unit, and if the environmental information used to calculate the potential occurrence probability includes information about a location that is a factor that increases the potential occurrence probability, the probability calculation unit calculates the potential occurrence probability to be lower as the distance between the location and the vehicle becomes longer. Another autonomous driving device disclosed herein is used in a vehicle capable of autonomous driving, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; a safety distance setting unit that sets a minimum safety distance from the vehicle that the vehicle should maintain between itself and an obstacle using at least information on the vehicle's behavior to avoid the vehicle coming into close proximity with the obstacle; and a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as the potential occurrence probability, which is the potential probability of the vehicle coming into close proximity with the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases.When increasing the safety distance set by the safety distance setting unit, the safety distance adjustment unit increases the safety distance so that it is greater than the shortest distance at which the vehicle can be stopped by autonomous driving, and less than the stopping distance that is the sum of the estimated free running distance and braking distance if the vehicle is stopped by manual driving. Another autonomous driving device disclosed herein is used in a vehicle capable of autonomous driving, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; a safety distance setting unit that uses at least information on the behavior of the vehicle to set a minimum safety distance from the vehicle that the vehicle should maintain between itself and an obstacle to avoid the vehicle coming into close proximity with the obstacle; a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as the potential occurrence probability, which is the potential probability of the vehicle coming into close proximity with the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases; and a response time adjustment unit that is capable of adjusting the response time from when an operation is instructed to the vehicle's braking device to operate to when the operation starts when the vehicle is stopped by autonomous driving, and the response time adjustment unit allows for a leeway in the response time when the potential occurrence probability is below a threshold, while when the potential occurrence probability is equal to or greater than the threshold, adjusts the response time to be shorter than when the potential occurrence probability is below the threshold. Another autonomous driving device disclosed herein is used in a vehicle capable of autonomous driving, and includes an environmental information acquisition unit that acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; a safety distance setting unit that sets a minimum safety distance from the vehicle that the vehicle should maintain between itself and an obstacle using at least information on the behavior of the vehicle to avoid the vehicle coming into close proximity with the obstacle; a safety distance adjustment unit that increases the safety distance set by the safety distance setting unit as the potential occurrence probability, which is the potential probability of the vehicle coming into close proximity with the obstacle based on the environmental information acquired by the environmental information acquisition unit, increases; and a lane changing unit that causes the vehicle to change lanes by autonomous driving, and when the potential occurrence probability is equal to or greater than a threshold, the road on which the vehicle is traveling is a multi-lane road on one side, and the lane on which the vehicle is traveling is not the lane of the multi-lane road with the lowest potential occurrence probability, the lane changing unit causes the vehicle to change lanes to a lane with a lower potential occurrence probability.

[0014] In order to achieve the above object, the autonomous driving method disclosed herein is used in an autonomous vehicle, acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle, and sets a minimum safety distance that the vehicle should keep between itself and an obstacle to avoid the vehicle coming into close contact with the obstacle, using at least information about the vehicle's behavior. The safety distance is increased in accordance with an increase in a potential occurrence probability, which is a potential occurrence probability of the vehicle coming into close contact with the obstacle based on the acquired environmental information. The safety distance is set at least in front of the vehicle, and the set safety distance in front of the vehicle is increased as the potential occurrence probability increases. The safety distance is also estimated in front of the vehicle following the vehicle as an obstacle, and when the following vehicle distance, which is the distance between the vehicle and the vehicle following the vehicle as an obstacle, is less than the estimated safety distance in front of the following vehicle, the distance by which the following vehicle distance falls short of the safety distance is added to the set safety distance in front of the vehicle. do. Another autonomous driving method disclosed herein is used in a vehicle capable of autonomous driving, and acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and sets the safety distance using at least information about the behavior of the vehicle. The minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle is increased as the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, increases. The safety distance is set at least on the left and right sides of the vehicle, and the number of moving objects detected around the vehicle is counted separately as moving objects to the left and right of the vehicle, and when the potential occurrence probability is equal to or greater than a threshold, only the safety distance on the left and right sides of the vehicle with the greater number of counted moving objects is increased. In addition, another autonomous driving method disclosed herein is used in a vehicle capable of autonomous driving, and acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and sets the minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle using at least information on the behavior of the vehicle. The minimum safety distance that the vehicle should maintain between itself and the obstacle to avoid close contact between the vehicle and the obstacle is increased as the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, increases, and the environmental information is used to calculate the potential occurrence probability.If the environmental information used to calculate the potential occurrence probability includes information about a location that is a factor that increases the potential occurrence probability, the potential occurrence probability is calculated to be lower as the distance between the location and the vehicle increases. In addition, another autonomous driving method disclosed herein is used in a vehicle capable of autonomous driving, and acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and sets the minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle using at least information about the vehicle's behavior.The minimum safety distance that the vehicle should maintain between itself and the obstacle to avoid close contact between the vehicle and the obstacle is increased as the potential probability of occurrence, which is the potential probability of close contact between the vehicle and the obstacle based on the acquired environmental information, increases.When increasing the safety distance, the safety distance is increased so that it is greater than the shortest distance at which the vehicle can be stopped by autonomous driving, and less than the stopping distance that is the sum of the estimated free running distance and braking distance if the vehicle is stopped by manual driving. In addition, another autonomous driving method disclosed herein is used in a vehicle capable of autonomous driving, and acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and sets the minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact with the vehicle using at least information on the behavior of the vehicle. The minimum safety distance that the vehicle should maintain between itself and the obstacle to avoid close contact with the vehicle is increased as the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, increases, and when stopping the vehicle by autonomous driving, the response time from when an operation is instructed to the vehicle's braking device to operate to when the operation starts is adjusted, and the response time is adjusted so that when the potential occurrence probability is below a threshold, a leeway is provided in the response time, while when the potential occurrence probability is equal to or greater than the threshold, the response time is adjusted to be shorter than when the potential occurrence probability is below the threshold. Another autonomous driving method disclosed herein is used in a vehicle capable of autonomous driving, and acquires environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and sets the minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle using at least information on the behavior of the vehicle. The minimum safety distance that the vehicle should maintain between itself and the obstacle to avoid close contact between the vehicle and the obstacle is increased as the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, increases, and the vehicle changes lanes through autonomous driving. When the potential occurrence probability during lane change is equal to or exceeds a threshold value, the road on which the vehicle is traveling is a multi-lane road in each direction, and the lane on which the vehicle is traveling is not the lane of the multi-lane road with the lowest potential occurrence probability, the vehicle is caused to change lanes to a lane with a lower potential occurrence probability. In order to achieve the above object, a control program disclosed herein is used in an autonomous vehicle, and causes a processor to acquire environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle, and to increase a minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid the vehicle coming into close contact with the obstacle, which is set using at least vehicle behavior information, in accordance with an increase in a potential occurrence probability, which is a potential occurrence probability of the vehicle coming into close contact with the obstacle based on the acquired environmental information. The safety distance is set at least ahead of the vehicle, and the set safety distance ahead of the vehicle is increased as the potential occurrence probability increases. The safety distance is also estimated ahead of the vehicle following the vehicle as an obstacle, and when the following vehicle distance, which is the distance between the vehicle and the vehicle following the vehicle as an obstacle, falls short of the estimated safety distance ahead of the following vehicle, the distance that the following vehicle distance falls short of the safety distance is added to the set safety distance ahead of the vehicle. . Another control program of the present disclosure is a control program used in a vehicle capable of autonomous driving, which causes a processor to acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and increase a safety distance, which is set using at least information on the behavior of the vehicle, that the vehicle should maintain between itself and an obstacle to avoid close contact with the obstacle, in accordance with an increase in the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, wherein the safety distance is set at least on the left and right sides of the vehicle, and the number of moving objects detected around the vehicle is counted separately as moving objects to the left and right of the vehicle, and when the potential occurrence probability is equal to or greater than a threshold, increase only the safety distance on the left and right sides of the vehicle with a larger number of counted moving objects. Another control program of the present disclosure is a control program used in a vehicle capable of autonomous driving, which causes a processor to acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and increase the minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle, which is set using at least information on the behavior of the vehicle, in accordance with an increase in the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information; the potential occurrence probability is calculated using the environmental information, and if the environmental information used to calculate the potential occurrence probability includes information about a location that is a factor that increases the potential occurrence probability, the potential occurrence probability is calculated to be lower in accordance with an increase in the distance between the location and the vehicle. Another control program of the present disclosure is a control program used in a vehicle capable of autonomous driving, which causes a processor to acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, and increase a safety distance, which is set using at least information on the behavior of the vehicle, that the vehicle should maintain as a minimum between the vehicle and an obstacle to avoid close contact between the vehicle and the obstacle, in accordance with an increase in the potential occurrence probability, which is the potential probability of close contact between the vehicle and the obstacle based on the acquired environmental information, and when increasing the safety distance, increase the safety distance so that it is greater than the shortest distance at which the vehicle can be stopped by autonomous driving, and less than the stopping distance that is the sum of the estimated free running distance and braking distance if the vehicle is stopped by manual driving. Another control program of the present disclosure is a control program used in a vehicle capable of autonomous driving, which causes a processor to acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, increase a minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact with the obstacle, which is set using at least information about the behavior of the vehicle, in accordance with an increase in the potential occurrence probability, which is the potential occurrence probability of close contact between the vehicle and the obstacle based on the acquired environmental information, and adjust the response time from when an operation is instructed to the vehicle's braking device to operate to when the operation starts when stopping the vehicle by autonomous driving, and the adjustment of the response time includes allowing a margin in the response time when the potential occurrence probability is below a threshold, while adjusting the response time to be shorter when the potential occurrence probability is equal to or greater than the threshold, compared to when the potential occurrence probability is below the threshold. Another control program of the present disclosure is a control program used in a vehicle capable of autonomous driving, which causes a processor to acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle, increase a minimum safety distance that the vehicle should maintain between itself and an obstacle to avoid close contact between the vehicle and the obstacle, which is set using at least information about the behavior of the vehicle, in accordance with an increase in the potential occurrence probability, which is the potential probability of close contact between the vehicle and the obstacle based on the acquired environmental information, and cause the vehicle to change lanes by autonomous driving, and when the potential occurrence probability is equal to or greater than a threshold value when the road on which the vehicle is traveling is a road with multiple lanes in each direction and the lane on which the vehicle is traveling is not the lane of the road with the lowest potential occurrence probability, the control program causes the vehicle to change lanes to a lane with a lower potential occurrence probability.

[0015] According to these, it is possible to increase the minimum safety distance that the vehicle should keep between the vehicle and the obstacle in order to avoid the vehicle coming close to the obstacle, as the potential occurrence probability, which is the potential occurrence probability of the vehicle coming close to the obstacle based on environmental information, which is information about the environment in which the vehicle is placed other than the obstacles around the vehicle, increases. Since this potential occurrence probability is based on environmental information, by increasing the safety distance as the potential occurrence probability increases, it is possible to increase the safety distance as the possibility of the vehicle coming close to the obstacle, which changes depending on the surrounding environment other than the vehicle's behavior, increases. If it becomes possible to increase the safety distance as the possibility of the vehicle coming close to the obstacle, which changes depending on the surrounding environment other than the vehicle's behavior, it becomes easier to avoid the vehicle coming close to the obstacle by the amount of the increase in the safety distance, even if the possibility of the vehicle coming close to the obstacle increases due to the surrounding environment other than the vehicle's behavior. As a result, it becomes easier to avoid the vehicle coming close to the obstacle during automatic driving of the vehicle. [Brief description of the drawings]

[0016] [Figure 1] 1 is a schematic configuration diagram of a vehicle system 1 and an automatic driving device 2. [Diagram 2] A diagram showing the relationship between factors and the likelihood of pedestrians running out into the road, divided by factor. [Diagram 3] 4 is a diagram for explaining an example of a safety distance in the forward and left and right directions of the host vehicle as a reference. FIG. [Figure 4] FIG. 13 is a graph showing the relationship between an increment in the safe distance and the ratio of the left and right count numbers. [Diagram 5] 11A and 11B are diagrams illustrating an example of increasing a safety distance ahead of the host vehicle. [Figure 6] FIG. 13 is a diagram for explaining a preferable range for increasing the safety distance. [Figure 7] 11 is a flowchart showing an example of the flow of a potential occurrence probability related process. [Figure 8] FIG. 13 is a configuration diagram of a vehicle system 801 according to an eighth embodiment. [Figure 9] FIG. 13 is a configuration diagram of a vehicle system 901 according to a ninth embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0017] A number of embodiments for disclosure will be described with reference to the drawings. For convenience of description, in a number of embodiments, parts having the same functions as parts shown in the drawings used in the previous description may be given the same reference numerals and their description may be omitted. For parts given the same reference numerals, the description in other embodiments may be referred to.

[0018] (Embodiment 1) <Overall configuration of vehicle system 1> Hereinafter, a first embodiment of the present disclosure will be described with reference to the drawings. A vehicle system 1 shown in FIG. 1 is used in a vehicle capable of autonomous driving (hereinafter, autonomous driving vehicle). As shown in FIG. 1, the vehicle system 1 includes an autonomous driving device 2, a locator 3, a map database (hereinafter, map DB) 4, a surrounding monitoring sensor 5, a vehicle control ECU 6, and a communication module 7. Although a vehicle using the vehicle system 1 is not necessarily limited to an automobile, the following description will be given taking an example of use in an automobile.

[0019] The autonomous vehicle in the first embodiment may be any vehicle capable of autonomous driving as described above. There may be multiple levels of autonomous driving (hereinafter, automation levels), as defined by, for example, the SAE. The automation levels are classified into levels Lv0 to Lv5 as follows, for example, according to the SAE definition.

[0020] Level Lv0 is a level where the driver performs all driving tasks without intervention of the system. Driving tasks are, for example, steering and acceleration / deceleration. Level Lv0 corresponds to so-called manual driving. Level Lv1 is a level where the system assists with either steering or acceleration / deceleration. Level Lv2 is a level where the system assists with both steering and acceleration / deceleration. Levels Lv1 to 2 correspond to so-called driving assistance.

[0021] Level 3 is a level where the system can perform all driving tasks in specific locations such as highways, and the driver performs driving operations in an emergency. At level 3, the driver is required to be able to respond quickly when the system requests a handover of driving. Level 3 corresponds to so-called conditional automated driving. Level 4 is a level where the system can perform all driving tasks except under specific circumstances such as unsupported roads and extreme environments. Level 4 corresponds to so-called highly automated driving. Level 5 is a level where the system can perform all driving tasks in any environment. Level 5 corresponds to so-called fully automated driving. Levels 3 to 5 correspond to so-called automated driving.

[0022] The autonomous vehicle in the first embodiment may be, for example, an autonomous vehicle with an automation level of level Lv3, or an autonomous vehicle with an automation level of level Lv4 or higher. The automation level may be switchable. As an example, the autonomous vehicle may be switchable between autonomous driving with an automation level of level Lv3 or higher and manual driving with level Lv0. Hereinafter, an example will be described in which an autonomous vehicle performs autonomous driving with an automation level of at least level Lv3 or higher.

[0023] The locator 3 includes a GNSS (Global Navigation Satellite System) receiver and an inertial sensor. The GNSS receiver receives positioning signals from a plurality of positioning satellites. The inertial sensor includes, for example, a gyro sensor and an acceleration sensor. The locator 3 sequentially determines the vehicle position of the vehicle equipped with the locator 3 by combining the positioning signals received by the GNSS receiver with the measurement results of the inertial sensor. The vehicle position is represented, for example, by latitude and longitude coordinates. Note that the vehicle position may be determined using a travel distance calculated from a signal sequentially output from a vehicle speed sensor equipped in the vehicle.

[0024] The map DB 4 is a non-volatile memory and stores map data such as link data, node data, road shapes, and structures. The map data may be a three-dimensional map consisting of a point cloud of characteristic points of road shapes and structures. When a three-dimensional map consisting of a point cloud of characteristic points of road shapes and structures is used as the map data, the locator 3 may be configured to specify the vehicle position using the three-dimensional map and a detection result by a perimeter monitoring sensor 5 such as a LIDAR (Light Detection and Ranging / Laser Imaging Detection and Ranging) or a perimeter monitoring camera that detects the point cloud of characteristic points of road shapes and structures, without using a GNSS receiver. The three-dimensional map may be generated based on captured images by REM (Road Experience Management).

[0025] The perimeter monitoring sensor 5 is an autonomous sensor that monitors the perimeter of the vehicle. As an example, the perimeter monitoring sensor 5 detects objects around the vehicle, such as moving objects (hereinafter, mobile objects) such as pedestrians, animals other than humans, and vehicles other than the vehicle, and stationary objects such as guardrails, curbs, and trees. In addition, the perimeter monitoring sensor 5 detects road markings such as lane markings around the vehicle. Examples of the perimeter monitoring sensor 5 include a perimeter monitoring camera that captures an image of a predetermined range around the vehicle, and a millimeter wave radar, sonar, LIDAR, and other distance measuring sensors that transmit search waves to a predetermined range around the vehicle. Note that a mobile object is an object that can move autonomously, and is not necessarily limited to a moving state, but may also include a stationary state.

[0026] The vehicle control ECU 6 is an electronic control device (i.e., a driving control device) that controls driving of the vehicle. Driving control includes acceleration / deceleration control and / or steering control. The vehicle control ECU 6 includes a steering ECU that controls steering, a power unit control ECU that controls acceleration / deceleration, and a brake ECU. The vehicle control ECU 6 controls driving by outputting control signals to each driving control device mounted on the vehicle, such as an electronically controlled throttle, a brake actuator, and an EPS (Electric Power Steering) motor.

[0027] The communication module 7 transmits and receives information to and from a center outside the vehicle via a public communication network. As an example, the communication module 7 downloads infrastructure information such as weather information, event information, and facility usage information from the center.

[0028] Weather information is information such as weather and temperature for each predetermined block unit. The block unit may be a map mesh unit, an administrative block unit, or another block unit. Event information is information about events that gather people, such as concerts. Event information may include the location of the event venue, the date the event is held, the time period of the event, etc. Facility usage information is information about the use of facilities where people gather, such as schools and train stations. Facility usage information may include the days of the week when the facility is used, the time period when the facility is used, etc. If the facility is a school, the time period when the facility is used may be from the time when the school arrives to the time when the school leaves. If the facility is a train station, the average number of users by time period may be included in the facility usage information.

[0029] The communication module 7 may be configured to receive the above-mentioned infrastructure information transmitted from a roadside device installed on the roadside through road-to-vehicle communication with the roadside device. The communication module 7 may receive map data distributed from an external server that distributes map data, and store the map data in the map DB 4.

[0030] The autonomous driving device 2 includes, for example, a processor, a memory, an I / O, and a bus connecting these, and executes a control program stored in the memory to execute processes related to autonomous driving. The memory referred to here is a non-transitory tangible storage medium that non-temporarily stores computer-readable programs and data. The non-transitory tangible storage medium is realized by a semiconductor memory or a magnetic disk. Details of the autonomous driving device 2 will be described below.

[0031] <Outline of the configuration of the automatic driving device 2> Next, a schematic configuration of the automatic driving device 2 will be described with reference to Fig. 1. As shown in Fig. 1, the automatic driving device 2 includes a vehicle position acquisition unit 21, a sensing information acquisition unit 22, an environmental information acquisition unit 23, a driving environment recognition unit 24, a probability calculation unit 25, and an automatic driving unit 26 as functional blocks. Note that some or all of the functions executed by the automatic driving device 2 may be configured as hardware using one or more ICs or the like. Also, some or all of the functional blocks included in the automatic driving device 2 may be realized by a combination of software execution by a processor and hardware members.

[0032] The vehicle position acquisition unit 21 acquires the vehicle position of the vehicle, which is sequentially measured by the locator 3. The sensing information acquisition unit 22 acquires the detection results (i.e., sensing information) sequentially detected by the surroundings monitoring sensor 5. The environment information acquisition unit 23 acquires information on the environment in which the vehicle is placed, other than obstacles around the vehicle (hereinafter, environment information). The environment information is information other than obstacles around the vehicle, and includes information on the frequency with which moving objects appear on the road. In addition, the information on obstacles around the vehicle, which is excluded from the environment information, means information obtained by detecting obstacles around the vehicle using a sensor provided in the vehicle. The content of the environment information itself includes information on the frequency with which moving objects appear on the road, in other words, information on the possibility that the position of the moving object will change from a non-road location to a road location over time. On the other hand, the change in content due to differences in the acquisition timing of the environment information is much smaller than the change in the position of the moving object over time, as can be seen from the above specific example. It can be said that the environment information is static information compared to the change in the position of the moving object over time.

[0033] The environmental information acquisition unit 23 includes a map information acquisition unit 231 and an infrastructure information acquisition unit 232 as sub-functional blocks. The map information acquisition unit 231 acquires map data stored in the map DB 4. The map data acquired by the map information acquisition unit 231 may include, for example, coordinates of facilities where people gather, such as schools and stations, longitudinal gradients of roads, and curvatures of curves. The map data acquired by the map information acquisition unit 231 corresponds to the above-mentioned environmental information. The map information acquisition unit 231 may acquire map data of the surroundings of the vehicle according to the vehicle position of the vehicle acquired by the vehicle position acquisition unit 21. It is preferable that the map information acquisition unit 231 acquires map data for a range wider than the detection range of the surroundings monitoring sensor 5.

[0034] The infrastructure information acquisition unit 232 acquires infrastructure information received from the center via the communication module 7. Examples of infrastructure information include the above-mentioned weather information, event information, facility usage information, etc. The infrastructure information acquired by the infrastructure information acquisition unit 232 corresponds to the above-mentioned environmental information.

[0035] The driving environment recognition unit 24 recognizes the driving environment of the vehicle from the vehicle position acquired by the vehicle position acquisition unit 21, the map data acquired by the map information acquisition unit 231, the sensing information acquired by the sensing information acquisition unit 22, and the like. As an example, the driving environment recognition unit 24 uses these pieces of information to recognize the positions, shapes, and moving states of objects around the vehicle, and generates a virtual space that reproduces the actual driving environment. The driving environment recognition unit 24 may also recognize the distance between the vehicle and the surrounding objects, the relative speed of the surrounding objects with respect to the vehicle, and the like as the driving environment from the sensing information acquired by the sensing information acquisition unit 22. In addition, when position information and speed information of surrounding vehicles can be acquired by vehicle-to-vehicle communication, the driving environment may be recognized using these pieces of information as well.

[0036] It is preferable that the driving environment recognition unit 24 distinguishes and recognizes whether the surrounding object detected by the periphery monitoring sensor 5 is a moving object or a stationary object. It is also preferable that the type of the surrounding object is also distinguished and recognized. The type of the surrounding object may be distinguished and recognized, for example, by performing pattern matching on an image captured by a periphery monitoring camera.

[0037] The types may be, for example, structures such as guardrails, objects fallen on the road, pedestrians, bicycles, motorcycles, automobiles, etc. Whether the surrounding objects are moving or stationary may be recognized according to the types of the surrounding objects. For example, if the type of the surrounding object is a structure or an object fallen on the road, it may be recognized as a stationary object. If the type of the surrounding object is a pedestrian, bicycle, motorcycle, or automobile, it may be recognized as a moving object. Note that objects that are unlikely to move immediately, such as parked vehicles, may be recognized as stationary objects. Parked vehicles may be recognized because they are stopped and their brake lights are not turned on by image recognition. As for the types of moving objects, for example, pedestrians may be further subdivided into children and adults. Children and adults may be distinguished by height, etc.

[0038] In addition, it is preferable that the driving environment recognition unit 24 also recognizes the behavior of the recognized moving object. As an example, it is sufficient to distinguish whether the moving object is stationary or moving from the presence or absence of change in the position of the moving object that is sequentially recognized along a time series and recognize it. In addition, the rate of change in the moving direction of the moving object may be recognized as the behavior of the moving object. As an example, the rate of change in the moving direction of the moving object may be calculated by identifying the moving direction of the moving object from the change in the position of the moving object that is sequentially recognized along a time series. It can be said that the larger the rate of change in the moving direction of the moving object, the more shaky the moving object is. This driving environment recognition unit 24 corresponds to the recognition unit.

[0039] The driving environment recognition unit 24 includes a counting unit 241 as a sub-functional block. The counting unit 241 counts (i.e., counts) the number of moving objects detected around the vehicle, distinguishing between moving objects on the left side of the vehicle and moving objects on the right side of the vehicle. This counting unit 241 corresponds to a counting unit. The counting unit 241 does not count objects recognized by the driving environment recognition unit 24 as stationary objects, but counts objects recognized as moving objects.

[0040] The counting unit 241 preferably counts the number of moving bodies by weighting the type of the moving body with a higher priority for protection in traffic. The priority for protection in traffic may be configured to be higher as the possibility of damage to the human body on the moving body increases when the vehicle collides with the moving body. Alternatively, the priority for protection in traffic may be configured to be higher as the vehicle's fault ratio in insurance increases when the vehicle collides with the moving body. As an example, the weighting for pedestrians may be configured to be higher than that for other types of moving bodies. Also, for example, the weighting according to the type may be pedestrian>bicycle>motorcycle>automobile. Note that, when the number of moving bodies is weighted and counted, the weighting may be configured to be multiplied by 1 and counted. The coefficient may be an integer or may not be an integer. If the coefficient is not an integer, the number of moving bodies counted will not be an integer.

[0041] The probability calculation unit 25 calculates a potential occurrence probability, which is a probability of occurrence of potential proximity between the host vehicle and an obstacle, by using the environmental information acquired by the environmental information acquisition unit 23. The potential occurrence probability is an example of an index representing the potential possibility of proximity between the host vehicle and an obstacle occurring. In this embodiment, the potential occurrence probability is used as an index representing the potential possibility of proximity between the host vehicle and an obstacle occurring. However, the above possibility may be expressed by a method of expression other than probability, such as by expressing it by a level such as high, medium, or low.

[0042] The potential occurrence probability may be limited to the potential occurrence probability of the proximity between the host vehicle and a pedestrian. The probability calculation unit 25 calculates the potential occurrence probability by using the environmental information based on the correlation between the environmental information and the potential occurrence probability.

[0043] For example, it is considered that the potential occurrence probability increases as the possibility P of a pedestrian running out increases. As shown in A of FIG. 2, it is considered that the possibility P of a pedestrian running out increases as the distance to a place where more people gather increases. Therefore, the probability calculation unit 25 may calculate the potential occurrence probability to be higher as the vehicle position of the host vehicle approaches the position of a facility where people gather, such as a school, a station, or an event venue. The positions of the school or station may be identified from map data acquired by the map information acquisition unit 231. The position of the event venue may be identified from event information acquired by the infrastructure information acquisition unit 232. The vehicle position of the host vehicle may be identified from the vehicle position of the host vehicle acquired by the vehicle position acquisition unit 21.

[0044] In this way, when the factor that increases the potential occurrence probability (hereinafter, the target factor) is a location, the probability calculation unit 25 calculates the potential occurrence probability to be lower as the distance between the target factor and the vehicle increases. In other words, when the environmental information used to calculate the potential occurrence probability includes information on a location that is a factor that increases the potential occurrence probability, the probability calculation unit 25 calculates the potential occurrence probability to be lower as the distance between the location and the vehicle increases.

[0045] In addition, the potential occurrence probability is considered to change depending on the time period, as shown in B and C of FIG. 2. For example, when the target factor is a school, the possibility P of pedestrians running out is considered to reach two peaks as shown in B of FIG. 2 during the time period between the time of arriving at school and the time of leaving school. For example, when the target factor is an event venue, the possibility P of pedestrians running out is considered to reach two peaks as shown in C of FIG. 2 before and after the time period of the event. For example, when the target factor is a station, the possibility P of pedestrians running out is considered to change depending on the average number of users by time period. Therefore, the probability calculation unit 25 may calculate the potential occurrence probability according to the time period for each target factor. The time of arriving at or leaving school and the average number of users by time period at the station may be specified from the facility usage information acquired by the infrastructure information acquisition unit 232. The end time of the event may be specified from the time period of the event in the event information acquired by the infrastructure information acquisition unit 232. In addition, when the target factor is a station, the possibility P of moving objects other than pedestrians running out is considered to be higher in the time period with a higher average number of users.

[0046] It is considered that the potential occurrence probability also changes depending on the weather. For example, when the weather is snowy, it is considered that the possibility that a moving object will not be able to stop and will jump out, or the braking distance of the vehicle will be extended, increases. Therefore, when the weather is snowy, the probability calculation unit 25 may calculate a higher potential occurrence probability than when the weather is not snowy. Also, when the weather is rainy, it is considered that the possibility P of a moving object other than a pedestrian jumping out into a facility such as a station for picking up or dropping off a person increases, or the braking distance of the vehicle will be extended. Therefore, when the weather is rainy, the probability calculation unit 25 may calculate a higher potential occurrence probability for a target factor such as a station than when the weather is not rainy.

[0047] It is considered that the potential occurrence probability also changes depending on the temperature. For example, when the temperature is low enough to freeze the road surface, it is considered that there is a high possibility that a moving object will not be able to stop and will jump out, or that the braking distance of the vehicle will be extended. Therefore, when the temperature is equal to or lower than the threshold temperature at which the road surface is estimated to freeze, the probability calculation unit 25 may calculate a higher potential occurrence probability than when the temperature is higher than the threshold temperature.

[0048] It is considered that the potential occurrence probability changes depending on the place where the vehicle is traveling. For example, it is considered that the more people there are in an area, the higher the possibility that pedestrians and bicycles will jump out. Therefore, the probability calculation unit 25 may calculate a higher potential occurrence probability as the area where the vehicle is traveling is more crowded. As an example, the types of areas may be classified in order of the number of people as follows: downtown > suburban road > expressway. Also, it is considered that the greater the curvature of the road, the higher the possibility that a motorcycle or automobile in the oncoming lane will run into the vehicle's lane. Therefore, the probability calculation unit 25 may calculate a higher potential occurrence probability as the road where the vehicle is traveling is more curved. In addition, it is considered that the braking distance of the vehicle will be extended as the longitudinal gradient of the downward direction increases. Therefore, the probability calculation unit 25 may calculate a higher potential occurrence probability as the longitudinal gradient of the lane where the vehicle is traveling increases.

[0049] The potential occurrence probability is considered to differ for each lane on a road with multiple lanes in each direction. For example, it is considered that the closer a lane is to the sidewalk, the higher the possibility that a pedestrian or bicycle will suddenly jump out. Therefore, when the road on which the vehicle is traveling is a road with multiple lanes in each direction, the closer the lane on which the vehicle is traveling is to the sidewalk, the higher the potential occurrence probability calculated by the probability calculation unit 25.

[0050] The probability calculation unit 25 may also be configured to calculate the potential occurrence probability by supplementing the sensing information acquired by the sensing information acquisition unit 22. The possibility P of a pedestrian jumping out is considered to change depending on the density and average height of pedestrians, as shown in D and E of FIG. 2. The density of pedestrians is the number of pedestrians per given range. For example, the probability calculation unit 25 may calculate the number of pedestrians per sensing range (i.e., density) and / or average height from the sensing information of moving bodies recognized as pedestrians by the driving environment recognition unit 24 among the surrounding objects detected by the perimeter monitoring sensor 5.

[0051] For example, as shown in D of FIG. 2, it is considered that the possibility of a pedestrian running out increases as the density of pedestrians increases. In addition, it is considered that the possibility of a pedestrian running out increases as the pedestrian is a child. Therefore, as shown in E of FIG. 2, it is considered that the possibility of a pedestrian running out increases as the average height of pedestrians decreases. Therefore, it is sufficient for the probability calculation unit 25 to calculate a high potential occurrence probability as the density of pedestrians increases. In addition, it is sufficient for the probability calculation unit 25 to calculate a high potential occurrence probability as the average height of pedestrians decreases.

[0052] In addition, the probability calculation unit 25 may calculate the potential occurrence probability by integrating a plurality of factors. For example, in the case of FIG. 2, when the target factor is a school, the potential occurrence probability may be calculated by integrating the likelihoods of A, B, D, and E in FIG. 2 as likelihoods LA, LB, LD, and LE, respectively. For example, the likelihood X1 obtained by integrating the likelihoods LA and LB is calculated by the formula X1=LA×LB / {LA×LB+(1-LA)×(1-LB)}. Next, the likelihood X2 obtained by integrating the likelihoods X1 and LD is calculated by the formula X2=X1×LD / {X1×LD+(1-X1)×(1-LD)}. Finally, the likelihood X3 obtained by integrating the likelihoods X2 and LE is calculated by the formula X3=X2×LE / {X2×LE+(1-X2)×(1-LE)}. Then, the obtained likelihood X3 is set as the potential occurrence probability obtained by integrating the likelihoods of A, B, D, and E. The combination of factors to be integrated is not limited to the above-mentioned examples.

[0053] The automatic driving unit 26 performs processing related to taking over driving operations from the driver. As shown in FIG. 1, the automatic driving unit 26 includes a driving plan unit 27, a confirmation unit 28, and an automatic driving function unit 29 as sub-function blocks.

[0054] The driving plan unit 27 generates a driving plan for driving the vehicle by automatic driving using the driving environment recognized by the driving environment recognition unit 24. For example, as a medium- to long-term driving plan, a route search process is performed to generate a recommended route for moving from the vehicle's position to the destination. In addition, as a short-term driving plan for driving according to the medium- to long-term driving plan, a driving plan for changing lanes, a driving plan for driving in the center of a lane, a driving plan for following a preceding vehicle, and a driving plan for avoiding obstacles are generated. The driving plan may be generated by the driving plan unit 27, for example, by calculating a route that is a certain distance or the center from the recognized lane marking, or by calculating a route that follows the behavior or driving trajectory of the recognized preceding vehicle.

[0055] Furthermore, the generation of the driving plan in the driving plan unit 27 may be configured to be performed by calculating a route determined to be optimal by machine learning or the like. When calculating a route determined to be optimal by machine learning, the route calculation algorithm for calculating the route includes a neural network. Examples of values ​​to be input to the neural network are the distance between the vehicle and surrounding vehicles, and the relative speed of the surrounding vehicles with respect to the vehicle. In addition, various driving environments recognized by the driving environment recognition unit 24 may be input to the neural network. The output of the neural network is the route.

[0056] This neural network is trained by machine learning such as reinforcement learning. In addition, multiple types of route calculation algorithms can be prepared. The differences between the multiple route calculation algorithms are differences in driving policies. A driving policy indicates the degree and frequency of acceleration / deceleration and steering control when automatically driving a vehicle. A driving policy is expressed, for example, as a difference in whether acceleration / deceleration is rapid or slow. An occupant of the vehicle can select which route calculation algorithm to use from multiple route calculation algorithms corresponding to multiple driving policies. When a route is calculated according to one selected route calculation algorithm, it can be said that the route is calculated based on a driving policy corresponding to that route calculation algorithm.

[0057] The driving plan unit 27 calculates one or more routes as a short-term driving plan. For example, the driving plan unit 27 may be configured to include, as the short-term driving plan, information on acceleration / deceleration for speed adjustment on the calculated route.

[0058] The confirmation unit 28 evaluates the safety of the driving plan generated by the driving plan unit 27. As an example, the confirmation unit 28 may evaluate the safety of the driving plan using a mathematical formula model that mathematically formulates the concept of safe driving in order to make it easier to evaluate the safety of the driving plan. The confirmation unit 28 may evaluate the safety based on whether or not the distance between the vehicle and a surrounding object (hereinafter, the inter-object distance) is equal to or greater than a reference distance (hereinafter, the safety distance) for evaluating the safety between the objects, which is calculated by a preset mathematical formula model. As an example, the inter-object distance may be the distance in the forward / rearward and / or left / right directions of the vehicle.

[0059] The mathematical formula model does not guarantee that an accident will not occur at all, but is intended to guarantee that if the vehicle takes appropriate action to avoid a collision when the safe distance is not reached, the vehicle will not be held responsible for the accident. The safe distance calculated by the mathematical formula model can be rephrased as the minimum distance that the vehicle should keep between itself and an obstacle in order to avoid the vehicle coming too close to the obstacle.

[0060] The confirmation unit 28 also includes a safety distance setting unit 281 and a safety distance adjustment unit 282 as sub-functional blocks, and provides a margin to the safety distance depending on the situation. More specifically, as the potential occurrence probability calculated by the probability calculation unit 25 increases, a margin is provided to the safety distance calculated using the mathematical formula model.

[0061] The safety distance setting unit 281 calculates the safety distance using the above-mentioned mathematical formula model, and sets the calculated safety distance as the safety distance. The safety distance setting unit 281 calculates and sets the safety distance using at least information on the behavior of the vehicle. The safety distance setting unit 281 may use, for example, an RSS (Responsibility Sensitive Safety) model as the mathematical formula model.

[0062] The safety distance setting unit 281 sets a safety distance in the forward and / or left / right directions of the host vehicle. In the following, an example will be described in which the safety distance setting unit 281 sets a safety distance in the forward and left / right directions of the host vehicle. When the safety distance setting unit 281 adopts a configuration in which it sets a safety distance in the forward or left / right directions of the host vehicle, it may be configured to perform the processing described below regarding the direction in which the safety distance is set.

[0063] The safety distance setting unit 281 may calculate, as a reference, a distance in front of the vehicle from the behavior information of the vehicle, which is, for example, a distance at which the vehicle can be stopped in the shortest time, as a safety distance. Taking FIG. 3 as an example, HV in FIG. 3 indicates the vehicle, and Fw indicates the front safe distance as a reference. As a specific example, the safety distance in front may be calculated as a distance at which the vehicle can decelerate and stop at the maximum deceleration after traveling forward at the maximum acceleration from the current vehicle speed to the response time from the speed, maximum acceleration, maximum deceleration, and response time of the vehicle. The speed, maximum acceleration, and maximum deceleration of the vehicle here are for the front-rear direction of the vehicle. The response time here may be the time from the instruction to operate the braking device to the start of the operation when the vehicle is stopped by automatic driving.

[0064] The safety distance setting unit 281 may set the forward safe distance as this reference even when it does not recognize a moving object in front of the host vehicle but recognizes a stationary object. When it recognizes a moving object in front of the host vehicle, the safety distance setting unit 281 may calculate the distance at which the host vehicle can stop without coming into contact with the moving object from information on the behavior of the host vehicle and the moving object as the forward safe distance.

[0065] The safety distance setting unit 281 may calculate, as a standard, a safe distance in the left-right direction of the vehicle from the behavior information of the vehicle, a distance that the vehicle moves in the left-right direction until the speed in the left-right direction can be reduced to 0 in the shortest time. For example, from the left-right speed, maximum acceleration, maximum deceleration, and response time of the vehicle, the distance that the vehicle moves in the left-right direction until the speed in the left-right direction can be reduced to 0 by decelerating at the maximum deceleration after moving in the left-right direction at the maximum acceleration during the response time from the current left-right speed may be calculated as the safe distance in the left-right direction. The response time here may be a time from an instruction to the steering device to operate to the start of the operation when steering the vehicle by automatic driving. The safety distance setting unit 281 may be configured to set the ratio of the left and right safety distances of the vehicle to be equal as a standard, as shown in FIG. 3. In FIG. 3, L indicates the safety distance to the left of the vehicle, and R indicates the safety distance to the right of the vehicle.

[0066] The safe distance setting unit 281 may set the safe distance in the left-right direction as this reference even when it recognizes a stationary object but does not recognize a moving object to the left or right of the own vehicle. When the safe distance setting unit 281 recognizes a moving object to the left or right of the own vehicle, it may calculate, for the direction in which the moving object exists, the distance that the own vehicle and the moving object move in the left or right direction until their speeds in the left and right direction become zero without coming into contact with each other, based on information on the behavior of the own vehicle and the moving object, as the safe distance in that direction. Therefore, when a moving object is recognized to the left or right of the own vehicle, the ratio of the safe distances to the left and right of the own vehicle may not be kept equal.

[0067] The safety distance setting unit 281 may also be configured to estimate a safe distance ahead for a vehicle following the own vehicle. In this case, the safe distance ahead for a vehicle following the own vehicle may be a distance at which, when the own vehicle at the current vehicle speed decelerates at the maximum deceleration, the following vehicle at the current vehicle speed travels forward at the maximum acceleration during the response time and then decelerates at the maximum deceleration without coming into contact with the own vehicle. The speed, maximum acceleration, and maximum deceleration of the following vehicle here are those for the front-rear direction of the following vehicle. The response time here may be the time from when an operation command is issued to the braking device to when the own vehicle is stopped by automatic driving to when the operation starts.

[0068] When the maximum acceleration, maximum deceleration, and response time of the following vehicle can be acquired from the following vehicle through vehicle-to-vehicle communication, the safety distance setting unit 281 may use the information acquired through vehicle-to-vehicle communication. The vehicle speed of the following vehicle may be that recognized by the driving environment recognition unit 24. Alternatively, the maximum acceleration, maximum deceleration, and response time of the following vehicle may be stored in advance in the non-volatile memory of the automatic driving device 2, and the safety distance setting unit 281 may use the values ​​of the following vehicle. The safety distance setting unit 281 may set the estimated safe distance ahead for the following vehicle as the safe distance behind the vehicle.

[0069] The safety distance adjustment unit 282 increases the safety distance set by the safety distance setting unit 281 in response to an increase in the potential occurrence probability calculated by the probability calculation unit 25. The increased safety distance may be the safety distance ahead of the host vehicle, or may be the safety distance in the left and right directions of the host vehicle.

[0070] For example, the safety distance adjustment unit 282 increases the safety distance ahead of the host vehicle set by the safety distance setting unit 281 in response to an increase in the potential occurrence probability calculated by the probability calculation unit 25. The correspondence relationship between the potential occurrence probability and the increase in the safety distance ahead of the host vehicle can be set arbitrarily.

[0071] According to the above configuration, it is possible to increase the safe distance ahead of the vehicle as the possibility of a moving object such as a pedestrian jumping out increases. Therefore, even if a moving object jumps out in front of the vehicle, it is possible to take evasive action to avoid approaching the moving object early. Therefore, it is possible to more easily avoid approaching an obstacle during automatic driving of the vehicle.

[0072] For example, the safety distance adjustment unit 282 may increase the safety distance in the left and right direction of the host vehicle set by the safety distance setting unit 281 in response to an increase in the potential occurrence probability calculated by the probability calculation unit 25. The correspondence relationship between the potential occurrence probability and the increase in the safety distance in the left and right direction of the host vehicle can be set arbitrarily.

[0073] According to the above configuration, it is possible to increase the safety distance in the left and right directions of the vehicle as the possibility of a moving object such as a pedestrian jumping out increases. Therefore, even if a moving object jumps out from the left or right direction of the vehicle, it is possible to take an avoidance action to avoid approaching the moving object early. Therefore, it is possible to more easily avoid approaching an obstacle during automatic driving of the vehicle. In addition, when the safety distance in the left and right directions of the vehicle is increased, the vehicle runs with a wider space between the vehicle and parked vehicles and structures on the left and right directions of the vehicle, so that the vehicle becomes easier to see from a moving object that is jumping out from behind the parked vehicles and structures. Therefore, it is possible to suppress the jumping out of the moving object by making it easier for the moving object to check the vehicle.

[0074] Furthermore, when the potential occurrence probability calculated by the probability calculation unit 25 is equal to or greater than a threshold, the safety distance adjustment unit 282 may increase only the safety distance on the left and right side of the vehicle set by the safety distance setting unit 281, the side on which the number of moving objects counted by the counting unit 241 is greater. The threshold here is a value for distinguishing whether or not there is a high possibility of a moving object jumping out, and may be any value that can be set. Note that, when no moving object is recognized to the left or right of the vehicle, the ratio of the equal left and right safety distances is changed by increasing the safety distance in either the left or right direction.

[0075] According to the above configuration, when the possibility of a moving object such as a pedestrian jumping out increases, it is possible to increase only the safety distance on the side where the number of moving objects is greater. Therefore, it is possible to increase the safety only on the left or right side of the vehicle where the possibility of a moving object jumping out is higher. If the safety distance on the left or right side of the vehicle is increased regardless of whether the possibility of a moving object jumping out is higher on the left or right side, there is a risk that the vehicle will take a useless avoidance action by increasing the safety distance in a useless direction. In contrast, by increasing the safety only on the left or right side of the vehicle where the possibility of a moving object jumping out is higher, it is possible to eliminate such waste and make it easier to avoid approaching an obstacle during automatic driving of the vehicle.

[0076] Furthermore, when increasing only the safety distance on the side where the number of moving objects counted by counting unit 241 is greater, the amount by which the safety distance is increased can be increased in accordance with an increase in the ratio of the numbers of moving objects on the left and right sides of the host vehicle counted by counting unit 241 (hereinafter, "count numbers"), as shown in Fig. 4. Fig. 4 is a diagram showing an example of the relationship between the increase in the safety distance and the ratio of the numbers of moving objects counted on the left and right sides of the host vehicle (hereinafter, "left / right ratio").

[0077] When the moving objects having the same number of counts are distributed on both the left and right sides of the vehicle, it is preferable to run near the center of the moving objects on the left and right sides in order to prepare for the moving objects jumping out from either the left or right direction, and it is preferable not to bias the left and right safety distances too much to one side. On the other hand, when the distribution is biased to one side of the vehicle, it is preferable to bias the safety distance to the side to which the distribution is biased in order to prepare for the moving objects jumping out from this side. In contrast, according to the above configuration, the amount by which the safety distance is increased is increased according to the increase in the ratio of the counts of the moving objects on the left and right sides of the vehicle, so that it is not necessary to uniformly increase the safety distance by the same amount when the moving objects having the same number of counts are distributed on both the left and right sides of the vehicle and when the distribution is biased to one side of the vehicle. Therefore, it is possible to more easily avoid approaching an obstacle according to the distribution of the moving objects on the left and right sides of the vehicle.

[0078] Furthermore, when the counting unit 241 is configured to count the number of moving bodies with a larger weighting for a type of moving body with a higher priority for protection in terms of traffic, it is possible to bias the safety distance on the left and right sides of the vehicle toward the side with a higher priority for protection in terms of traffic, out of the left and right sides of the vehicle. Therefore, it becomes easier to avoid approaching an obstacle on the side with a higher priority for protection in terms of traffic, out of the left and right sides of the vehicle.

[0079] When increasing the safety distance set by the safety distance setting unit 281, the safety distance adjustment unit 282 may change the increase rate of the safety distance according to the type of moving object recognized by the traveling environment recognition unit 24. For example, the higher the type of moving object is estimated to be the type with a higher possibility of jumping out, the higher the increase rate of the safety distance may be. In this case, the increase rate of the safety distance may be changed according to the type of moving object estimated to be the most likely to jump out among the moving objects recognized by the traveling environment recognition unit 24. According to this, the higher the possibility of the recognized moving object jumping out is, the more the safety distance is increased, making it easier to avoid approaching an obstacle. As an example, when the type of moving object is a child, the increase rate of the safety distance may be changed to be higher than when the type of moving object is an adult.

[0080] Also, the higher the priority of a type of moving object to be protected from traffic, the higher the increase rate of the safety distance may be. As an example, the increase rate of the safety distance may be changed according to the type of moving object that has the highest priority of being protected from traffic among the moving objects that can be recognized by the driving environment recognition unit 24. In this way, the higher the priority of a recognized moving object to be protected from traffic, the more the safety distance is increased, making it easier to avoid approaching a moving object with a higher priority of being protected from traffic.

[0081] When the safety distance adjustment unit 282 increases the safety distance set by the safety distance setting unit 281, the safety distance adjustment unit 282 may change the increase rate of the safety distance according to the behavior of the moving object recognized by the traveling environment recognition unit 24. For example, the more the behavior of the moving object is estimated to be likely to jump out, the higher the increase rate of the safety distance may be. In this case, the increase rate of the safety distance may be changed according to the behavior of the moving object that is estimated to be most likely to jump out among the moving objects that can be recognized by the traveling environment recognition unit 24. According to this, the higher the possibility of the recognized moving object jumping out, the more the safety distance is increased, making it easier to avoid approaching an obstacle. As an example, when the behavior of the moving object corresponds to moving, the increase rate of the safety distance may be changed to be higher than when the behavior of the moving object is stationary. Also, the increase rate of the safety distance may be changed to be higher according to the increase in the rate of change in the moving direction of the moving object as the behavior of the moving object. In other words, the increase rate of the safety distance may be changed to be higher according to the increase in the wobbling of the moving object.

[0082] As shown in Fig. 5, when the distance between the host vehicle and the following vehicle (hereinafter, the following vehicle distance) is less than the safety distance ahead of the following vehicle estimated by the safety distance setting unit 281, the safety distance adjustment unit 282 may be configured to increase the distance by which the following vehicle distance falls short of the safety distance by adding it to the safety distance ahead of the host vehicle set by the safety distance setting unit 281. In Fig. 5, HV indicates the host vehicle, SV indicates the following vehicle, Fw indicates the safety distance ahead of the host vehicle, and Re indicates the distance equivalent to the safety distance ahead of the following vehicle. Also, in Fig. 5, Exa indicates the distance by which the following vehicle distance falls short of the safety distance ahead of the following vehicle. Note that the following vehicle distance may be configured to be recognized by the driving environment recognition unit 24.

[0083] According to the above configuration, the insufficient safety distance ahead of the following vehicle is added to the safety distance ahead of the own vehicle. Here, a case will be described in which the own vehicle stops when the distance between the own vehicle and the obstacle ahead of the own vehicle becomes the safety distance, and the following vehicle also stops. If the safety distance ahead of the own vehicle is increased, the own vehicle can stop at a distance equal to the safety distance without decelerating at the maximum deceleration. Since the safety distance ahead of the following vehicle is a value assuming that the own vehicle decelerates at the maximum deceleration, the distance that the following vehicle can travel without contacting the own vehicle is extended by the amount by which the own vehicle decelerates at a deceleration lower than the maximum deceleration. Here, the amount by which the safety distance ahead of the own vehicle is increased is the insufficient safety distance ahead of the following vehicle, so the following vehicle can stop without contacting the own vehicle due to this extended distance. Therefore, in this respect, it becomes easier to avoid approaching an obstacle during automatic driving of the vehicle.

[0084] In addition, when the safety distance adjustment unit 282 increases the safety distance set by the safety distance setting unit 281, it is preferable to increase the safety distance so that it is within a range of the shortest distance (see AD in FIG. 6) at which the vehicle can be stopped by automatic driving and less than the stopping distance (see MD in FIG. 6) that is the sum of the free running distance and the braking distance estimated when the vehicle is stopped by manual driving, as shown in FIG. 6. In other words, the distance obtained by subtracting the safety distance set by the safety distance setting unit 281 from the stopping distance may be set as the upper limit value when increasing the safety distance. In FIG. 6, Fw indicates the safety distance ahead of the vehicle, and Exb indicates the distance by which the safety distance is increased. As for the stopping distance, a correspondence relationship between the stopping distance and the speed of the vehicle may be stored in advance in the non-volatile memory of the automatic driving device 2, and the stopping distance may be specified and used by the safety distance adjustment unit 282 based on this correspondence relationship and the vehicle speed of the vehicle.

[0085] According to the above configuration, even when the safety distance is increased, by setting the safety distance to be less than the stopping distance when driving manually, it is possible to take evasive action more quickly than when driving manually, and even if the accident cannot be avoided, it is possible to claim that the accident could not have been avoided even with manual driving.

[0086] The confirmation unit 28 may evaluate the driving plan generated by the driving plan unit 27 as safe when the inter-object distance is equal to or greater than the safety distance set by the safety distance setting unit 281. On the other hand, the confirmation unit 28 may evaluate the driving plan generated by the driving plan unit 27 as unsafe when the inter-object distance is less than this safety distance. The confirmation unit 28 may output the driving plan evaluated as safe to the automatic driving function unit 29. On the other hand, the confirmation unit 28 may correct the driving plan evaluated as unsafe, for example, to a driving plan evaluated as safe, and output it to the automatic driving function unit 29. In other words, whether or not to use the route calculated by the driving plan unit 27 for automatic driving is evaluated depending on whether or not the inter-object distance is equal to or greater than the safety distance.

[0087] The automatic driving function unit 29 may perform driving operations on behalf of the driver (i.e., automatic driving) by having the vehicle control ECU 6 automatically accelerate / decelerate and / or steer the vehicle in accordance with the driving plan output from the confirmation unit 28. The automatic driving function unit 29 performs automatic driving along the route evaluated by the confirmation unit 28 to be used for automatic driving. The automatic driving function unit 29 performs automatic driving in accordance with the driving plan output from the confirmation unit 28, thereby performing evasive action to avoid the vehicle coming into close contact with surrounding objects.

[0088] Furthermore, the automatic driving function unit 29 includes a response time adjustment unit 291 and a lane change unit 292 as sub-function blocks. The response time adjustment unit 291 adjusts the response time from when an instruction to operate is given to the braking device to when the vehicle starts operating when the vehicle is stopped by automatic driving. As an example, the braking device is provided with a certain margin (hereinafter, "play time") other than the minimum time that cannot be shortened as a default during automatic driving in this response time. The response time adjustment unit 291 may perform control to change this play time of the braking device.

[0089] When the potential occurrence probability calculated by the probability calculation unit 25 is less than the aforementioned threshold value, the response time adjustment unit 291 provides a margin in the response time, whereas when the potential occurrence probability calculated by the probability calculation unit 25 is equal to or greater than the threshold value, it is preferable to adjust the response time to be shorter than when the potential occurrence probability calculated by the probability calculation unit 25 is less than the threshold value.

[0090] As an example, the response time adjustment unit 291 may be configured to change the idle time to be shorter than the default when the potential occurrence probability is less than a threshold. This makes it possible to shorten the response time and make it easier to avoid approaching an obstacle when the possibility of a moving object jumping out increases. Since the shorter the response time, the easier it is to avoid approaching an obstacle, it is more preferable for the response time adjustment unit 291 to change the idle time to 0 when the potential occurrence probability is less than a threshold.

[0091] The lane changing unit 292 automatically drives the host vehicle to change lanes. The lane changing unit 292 may automatically drive the host vehicle to change lanes in accordance with the driving plan for lane change generated by the driving plan unit 27. As an example, the driving plan unit 27 may generate a driving plan for changing lanes to a lane with a lower potential occurrence probability when the potential occurrence probability calculated by the probability calculation unit 25 is equal to or higher than the threshold value, the road on which the host vehicle is traveling is a multi-lane road in one direction, and the lane on which the host vehicle is traveling is not the lane with the lowest potential occurrence probability calculated by the probability calculation unit 25 among the multi-lane road in one direction. As a result, the lane changing unit 292 changes lanes to a lane with a lower potential occurrence probability under the above-mentioned conditions. Note that the lane to be changed to is limited to the lane in the same link direction as the lane on which the host vehicle is traveling.

[0092] According to the above configuration, the possibility of a moving object jumping out is reduced by changing lanes to a lane where the possibility of a moving object jumping out is lower. Therefore, it becomes easier to avoid approaching an obstacle by the amount of the reduction in the possibility of the moving object jumping out.

[0093] <Processing related to potential occurrence probability in autonomous driving device 2> Here, an example of the flow of processing according to the potential occurrence probability in the automatic driving device 2 (hereinafter, potential occurrence probability related processing) will be described using the flowchart of FIG. 7. Execution of steps included in the potential occurrence probability related processing by a computer corresponds to execution of an automatic driving method. The flowchart of FIG. 7 may be configured to start when a switch (hereinafter, power switch) for starting the internal combustion engine or motor generator of the vehicle is turned on and automatic driving is started. Alternatively, in the case of a configuration in which the vehicle can be switched between manual driving and automatic driving, the flowchart may be configured to start when the power switch is turned on while the vehicle is set to perform automatic driving. Alternatively, the flowchart may be configured to start when the setting for performing automatic driving is switched on during manual driving and the vehicle is switched to automatic driving.

[0094] First, in step S1, various information is acquired. In S1, the vehicle position acquisition unit 21 acquires the vehicle position of the vehicle. In S1, the sensing information acquisition unit 22 acquires sensing information. In S1, the environmental information acquisition unit 23 acquires environmental information. As the environmental information, the map information acquisition unit 231 acquires map data, and the infrastructure information acquisition unit 232 acquires infrastructure information.

[0095] In step S2, the driving environment recognition unit 24 recognizes the driving environment of the vehicle. In step S3, the counting unit 241 counts the number of moving objects detected around the vehicle by weighting the number of moving objects according to the type of the moving objects, and distinguishes between moving objects on the left side and moving objects on the right side of the vehicle.

[0096] In step S4, the safety distance setting unit 281 calculates the safety distance using a mathematical formula model, and sets the calculated safety distance as the safety distance. Here, the safety distance in front of the host vehicle and the safety distance in the left and right directions of the host vehicle are set. Also, in S4, if a moving object is recognized as an object in the vicinity of the host vehicle in S2, the safety distance in the recognized direction of the moving object is calculated using information on the behavior of the moving object as well. If a vehicle following the host vehicle is recognized in S2, the safety distance setting unit 281 also estimates the safety distance in front of the following vehicle.

[0097] In step S5, the probability calculation unit 25 calculates the potential occurrence probability using the environmental information acquired in S1. In step S6, the safety distance adjustment unit 282 increases the safe distance ahead of the host vehicle set in S4 as the potential occurrence probability calculated in S5 increases.

[0098] In step S7, if the vehicle following the host vehicle was recognized in S2 and the distance between the vehicle and the following vehicle is less than the safety distance ahead of the following vehicle estimated in S4, that is, if the safety distance ahead of the following vehicle is insufficient (YES in S7), the process proceeds to step S8. On the other hand, if the safety distance ahead of the following vehicle is not insufficient (NO in S7), the process proceeds to step S9. In step S8, safety distance adjustment unit 282 adds the distance by which the following vehicle distance is less than the safety distance ahead of the following vehicle to the safety distance ahead of the host vehicle set in S4.

[0099] In step S9, if the potential occurrence probability calculated in S5 is equal to or greater than the threshold (YES in S9), the process proceeds to step S10. On the other hand, if the potential occurrence probability calculated in S5 is less than the threshold (NO in S9), the process proceeds to step S12.

[0100] In step S10, the safety distance adjustment unit 282 increases only the safety distance on the side where the number of moving objects counted in S3 is greater, out of the left and right safety distances set in S4. In step S11, the response time adjustment unit 291 determines that there is no idle time to be provided in the response time, and the process proceeds to step S13. On the other hand, in step S12, the response time adjustment unit 291 determines that there is idle time to be provided in the response time.

[0101] In step S13, if the road on which the host vehicle is traveling is a multi-lane road in each direction and there is a lane with a lower potential occurrence probability calculated by the probability calculation unit 25 than the lane on which the host vehicle is traveling (YES in S13), the process proceeds to step S14. On the other hand, if there is no lane with a lower potential occurrence probability calculated by the probability calculation unit 25 than the lane on which the host vehicle is traveling (NO in S13), the process proceeds to step S15. In step S14, the lane changing unit 292 causes the host vehicle to change lanes by automatic driving to a lane with a lower potential occurrence probability calculated by the probability calculation unit 25 than the lane on which the host vehicle is traveling.

[0102] In step S15, if it is time to end the potential occurrence probability related process (YES in S15), the potential occurrence probability related process is ended. On the other hand, if it is not time to end the potential occurrence probability related process (NO in S15), the process returns to S1 and is repeated. Examples of timings to end the potential occurrence probability related process include when the power switch of the vehicle is turned off, when the vehicle is switched to manual driving, etc. Note that the flowchart in FIG. 7 is merely an example, and the order of the processes may be partially changed.

[0103] <Summary of the First Embodiment> According to the configuration of the first embodiment, the possibility of approaching an obstacle that changes depending on the surrounding environment other than the behavior of the vehicle is expressed by the potential occurrence probability. Since the safety distance set using at least the information on the behavior of the vehicle is increased as the potential occurrence probability increases, it becomes possible to increase the safety distance as the possibility of approaching an obstacle that changes depending on the surrounding environment other than the behavior of the vehicle increases. Therefore, even if the possibility of approaching an obstacle increases due to factors of the surrounding environment other than the behavior of the vehicle, it becomes easier to avoid approaching the obstacle by the amount of the increase in the safety distance. As a result, it becomes easier to avoid approaching an obstacle during automatic driving of the vehicle.

[0104] (Embodiment 2) In the first embodiment, the response time adjustment unit 291 adjusts the idle time provided in the response time, but this is not necessarily limited to the above. For example, the automatic driving function unit 29 may not include the response time adjustment unit 291, and the idle time provided in the response time may not be adjusted.

[0105] (Embodiment 3) In the first embodiment, the lane changing unit 292 automatically changes the lane of the host vehicle to a lane having a lower potential occurrence probability calculated by the probability calculation unit 25 than the lane in which the host vehicle is traveling, but the present invention is not necessarily limited to this. For example, the lane may not be changed according to the potential occurrence probability for each lane of a road with multiple lanes in each direction.

[0106] (Embodiment 4) In the first embodiment, the configuration has been shown in which, of the left and right safety distances set by the safety distance setting unit 281, only the safety distance on the side where the number of moving objects counted by the counting unit 241 is greater is increased, but this is not necessarily limited to this. For example, the counting unit 241 may not be provided in the traveling environment recognition unit 24, and each of the left and right safety distances set by the safety distance setting unit 281 may be increased as the potential occurrence probability calculated by the probability calculation unit 25 becomes higher.

[0107] (Embodiment 5) In the first embodiment, the probability calculation unit 25 calculates the potential occurrence probability using both the map data acquired by the map information acquisition unit 231 and the infrastructure information acquired by the infrastructure information acquisition unit 232, but this is not necessarily limited to this. For example, the probability calculation unit 25 may calculate the potential occurrence probability using only one of the map data acquired by the map information acquisition unit 231 and the infrastructure information acquired by the infrastructure information acquisition unit 232.

[0108] (Embodiment 6) In the first embodiment, the probability calculation unit 25 calculates the potential occurrence probability, but the present invention is not limited to this. For example, the potential occurrence probability may not be calculated, and the safety distance may be increased according to the potential occurrence probability by distinguishing between situations with different potential occurrence probabilities. Examples of situations with different potential occurrence probabilities include a case where the potential occurrence probability is equal to or greater than the threshold value and a case where the potential occurrence probability is less than the threshold value. The situations with different potential occurrence probabilities may be divided into three or more stages. When distinguishing between situations with different potential occurrence probabilities without calculating the potential occurrence probability, the configuration may be such that the situations with different potential occurrence probabilities are distinguished and specified based on the environmental information acquired by the environmental information acquisition unit 23 by referring to a table, map, or the like in which combinations of environmental information acquired by the environmental information acquisition unit 23 are associated with situations with different potential occurrence probabilities.

[0109] According to the configuration of the sixth embodiment, the safety distance from the vehicle to the obstacle, which is the minimum distance that the vehicle should keep between the vehicle and the obstacle in order to avoid the vehicle approaching the obstacle, can be increased as the potential occurrence probability, which is the potential occurrence probability of the vehicle approaching the obstacle based on the environmental information, which is information on the environment in which the vehicle is placed other than the obstacle around the vehicle, increases. Since this potential occurrence probability is based on the environmental information, by increasing the safety distance as the potential occurrence probability increases, the safety distance can be increased as the possibility of the vehicle approaching the obstacle, which changes depending on the surrounding environment other than the vehicle's behavior, increases. If the safety distance can be increased as the possibility of the vehicle approaching the obstacle, which changes depending on the surrounding environment other than the vehicle's behavior, increases, the vehicle can avoid approaching the obstacle by the increased safety distance, even if the possibility of the vehicle approaching the obstacle increases due to the surrounding environment other than the vehicle's behavior. As a result, the vehicle can avoid approaching the obstacle more easily during automatic driving.

[0110] (Embodiment 7) In the first embodiment, the default safety distance is calculated by a mathematical formula model, but the present invention is not limited to this. For example, the default safety distance may be calculated by a formula other than a mathematical formula model. For example, the safety distance setting unit 281 may calculate the safety distance by using information on the behavior of the vehicle and the mobile objects around the vehicle using other indices such as TTC (Time To Collision).

[0111] (Embodiment 8) Next, an eighth embodiment will be described. Fig. 8 shows a configuration diagram of a vehicle system 801 according to the eighth embodiment. The vehicle system 801 according to the eighth embodiment includes an automatic driving device 802, a behavior detection sensor 805, and a rule DB storage unit 806 as configurations different from those of the vehicle system 1 according to the first embodiment.

[0112] The behavior detection sensor 805 is a sensor group having a plurality of sensors, and includes the periphery monitoring sensor 5 described in the first embodiment, and further includes a sensor for detecting the behavior of the host vehicle. The periphery monitoring sensor 5 can detect the behavior of moving objects existing around the host vehicle, and non-moving obstacles existing around the host vehicle. The sensor for detecting the behavior of the host vehicle includes a sensor for detecting the speed and acceleration of the host vehicle, and a sensor for detecting the moving direction of the host vehicle. Note that, even in the first embodiment, the automatic driving device 2 may acquire detection values ​​from the sensor for detecting the behavior of the host vehicle.

[0113] The area surrounding the vehicle may be, for example, a rectangular area with the vehicle at the center and parallel sides in the front-rear and left-right directions of the vehicle. The size of the rectangle in front of the vehicle may be approximately the stopping distance of the vehicle. The distance behind the vehicle may be the same as the front direction, or may be shorter. The size of the rectangle in the left-right directions of the vehicle may be the length of one or several lanes. The size of the surrounding area may be set in various ways. The shape of the surrounding area may also be set in various ways. For example, the shape of the surrounding area may be a perfect circle or an ellipse.

[0114] The rule DB storage unit 806 is a storage unit that stores a rule database (hereinafter, rule DB). The rule DB is a database that stores rules (hereinafter, accident responsibility rules) that serve as standards for determining responsibility for potential accidents, which will be described later. This rule DB stores accident responsibility rules by location. The accident responsibility rules by location include traffic directions such as one-way streets, speed limits, and priority / non-priority distinctions. Of course, the accident responsibility rules also include rules that are not dependent on location. The accident responsibility rules that are not dependent on location include rules regarding the correspondence between traffic light colors and stopping driving, and giving priority to pedestrians over vehicles.

[0115] The automatic driving device 802 is a device that automatically drives a vehicle, and is an example of a vehicle control device. The automatic driving device 802 executes an automatic driving method for automatically driving the vehicle, similar to the automatic driving device 2. The automatic driving method is an example of a vehicle control method.

[0116] The automatic driving device 802 has a part of a configuration different from the automatic driving device 2 provided in the vehicle system 1 of the embodiment 1. Like the automatic driving device 2, the automatic driving device 802 includes a vehicle position acquisition unit 21, a sensing information acquisition unit 22, an environmental information acquisition unit 23, and a probability calculation unit 25. In addition, the automatic driving device 802 includes a driving environment recognition unit 824, an automatic driving unit 826, and a storage medium 840.

[0117] The traveling environment recognition unit 824 is the same as the traveling environment recognition unit 24 of the first embodiment in that it has the count unit 241 described in the first embodiment as a sub-function. Moreover, the traveling environment recognition unit 824 has all the functions of the traveling environment recognition unit 24 of the first embodiment. In addition, the traveling environment recognition unit 824 also recognizes the current behavior of the host vehicle. The behavior of the host vehicle is determined based on the detection value of a sensor that detects the behavior of the host vehicle, which is included in the behavior detection sensor 805. The behavior of the host vehicle can be expressed by a predetermined index including, for example, the current speed, acceleration, and yaw rate of the host vehicle.

[0118] The autonomous driving unit 826 includes a behavior prediction unit 828, a scene determination unit 829, and a responsibility determination unit 830 in addition to the driving plan unit 27, the confirmation unit 28, and the autonomous driving function unit 29 described in the first embodiment.

[0119] The behavior prediction unit 828 predicts the behavior of the prediction target moving object. The prediction target moving object is a concept including a vehicle (i.e., the vehicle) on which the vehicle system 801 is mounted and at least one moving object existing around the vehicle. Hereinafter, a moving object existing around the vehicle is referred to as a surrounding moving object. Examples of surrounding moving objects are the above-mentioned surrounding vehicles and pedestrians. The behavior prediction unit 828 predicts the behavior of the prediction target moving object for a certain period of time set in advance. The certain period of time is, for example, equal to or slightly longer than the time required to travel the candidate route. The behavior of the prediction target moving object is expressed including the elapsed time from the current time and the position at that time. The elapsed time expressing the behavior of the prediction target moving object is a discrete value with the unit elapsed time being one unit. The unit elapsed time is several milliseconds to several tens of milliseconds, depending on the processing power of the device.

[0120] In order to predict the behavior of the prediction target moving object, the behavior prediction unit 828 acquires the recognition result of the driving environment recognition unit 824. The recognition result of the driving environment recognition unit 824 includes the behavior of surrounding vehicles, the behavior of moving objects other than surrounding vehicles that exist around the vehicle, and stationary obstacles around the vehicle.

[0121] The behavior prediction unit 828 also acquires a potential occurrence probability, which is an example of an index representing the potential possibility of the host vehicle coming close to an obstacle, from the probability calculation unit 25. The behavior prediction unit 828 then predicts the behavior of the prediction target moving object by taking this potential occurrence probability into consideration. For example, the behavior of the host vehicle, which is the prediction target moving object, is predicted to limit the speed as the potential occurrence probability increases. The behavior of the host vehicle may also be predicted to increase the inter-vehicle distance as the potential occurrence probability increases.

[0122] The behavior prediction unit 828 can also predict the behavior of the prediction target moving object with a probability distribution. One example of the probability distribution is a probability distribution of the location of the prediction target moving object at each time when the behavior of the prediction target moving object is predicted. The higher the potential occurrence probability, the wider the probability distribution may be for the behavior prediction of the prediction target moving object.

[0123] When the prediction target moving object is a nearby vehicle, predictions may be made in the same way as for the own vehicle, such that the higher the potential occurrence probability, the greater the speed, the longer the inter-vehicle distance, and the wider the probability distribution.

[0124] Furthermore, in this embodiment, the behavior prediction unit 828 also acquires the route calculated by the driving plan unit 27 as a short-term driving plan. In the following embodiment, the route calculated by the driving plan unit 27 as a short-term driving plan is considered to be a candidate route. The reason for calling it a "candidate" route is that the route calculated as a short-term driving plan is a candidate route to be used for automatic driving, and may not be used for automatic driving depending on the judgment result of the responsibility judgment unit 830. The candidate route is expressed including the time at which the vehicle passes each position on the route in addition to the position where the vehicle is traveling. Even if the position is the same, the distance to the surrounding moving object is different if the time is different, so even if the position is the same, the possibility of an accident is different if the time is different.

[0125] As an example of the behavior of a prediction target moving object predicted by the behavior prediction unit 828, the behavior predicted for the own vehicle will be described. First, the position of the own vehicle when a unit elapsed time has elapsed is predicted based on the current behavior of the own vehicle. Then, the behavior of the own vehicle is predicted assuming that the vehicle control ECU 6 controls the behavior of the own vehicle so that the vehicle travels along the candidate route acquired from the travel planning unit 27. However, the behavior of the own vehicle is predicted by predicting whether or not the vehicle will be unable to travel along the candidate route due to the travel environment, such as the need to change to a route that avoids surrounding moving objects.

[0126] The behavior of the surrounding vehicles is predicted based on the assumption that the surrounding vehicles are controlled to travel along the planned route, which is successively transmitted by wireless, and when the planned route can be acquired via the communication module 7, the surrounding vehicles are controlled to travel along that route. However, the behavior of the surrounding vehicles is predicted based on the assumption that the surrounding vehicles are controlled to travel along the planned route, based on the obstacles present around the surrounding vehicles.

[0127] In the case of a prediction target moving body whose planned movement route cannot be obtained, such as a nearby vehicle or pedestrian, the behavior of the prediction target moving body is predicted based on the current movement speed and movement direction of the prediction target moving body.

[0128] The scene determination unit 829 acquires accident responsibility rules at the current vehicle position and at each position on the candidate route. In other words, the scene determination unit 829 determines what accident responsibility rule should be adopted at the current vehicle position and at each position on the candidate route.

[0129] Accident liability rules include accident liability rules that vary depending on the location and accident liability rules that are not dependent on the location. An example of an accident liability rule that varies depending on the location is the priority relationship at an intersecting road. This priority relationship is determined from the road shape or road signs. Another example of an accident liability rule that varies depending on the location is traffic rules that are determined for each location. An example of a traffic rule that is determined for each location is a speed limit. An example of a rule that is not dependent on the location is a required following distance that is determined depending on the speed.

[0130] In order to obtain different accident responsibility rules depending on the locations at various points indicated by the candidate route, the scene determination unit 829 obtains the candidate route. However, if different accident responsibility rules depending on the locations are obtained taking into consideration the length of the candidate route, the scene determination unit 829 does not need to obtain the candidate route.

[0131] The responsibility determination unit 830 determines the responsibility of the own vehicle for a potential accident that is assumed when the own vehicle travels along the candidate route, based on the behavior predicted by the behavior prediction unit 828. To make this determination, the responsibility determination unit 830 acquires the behavior of the prediction target moving object predicted by the behavior prediction unit 828, and acquires the accident responsibility rule acquired by the scene determination unit 829 from the scene determination unit 829. Then, the responsibility determination unit 830 determines the responsibility of the own vehicle for a potential accident that is assumed when the own vehicle travels along the candidate route, by applying the behavior of the prediction target moving object predicted by the behavior prediction unit 828 to the accident responsibility rule acquired by the scene determination unit 829.

[0132] This responsibility is determined based on an index of whether or not the safe distance described in the first embodiment is secured between the vehicle and the obstacle. Therefore, the responsibility determination unit 830 includes the confirmation unit 28 described in the first embodiment.

[0133] Responsibility for potential accidents can be expressed as a value (hereafter referred to as potential accident responsibility value). If the potential accident responsibility value of the vehicle is α%, the degree of responsibility of surrounding moving objects that are assumed to have a potential accident with the vehicle can also be expressed as a value such as 100-α(%).

[0134] The potential accident responsibility value can be calculated from a previously set formula, such as the mathematical formula model disclosed in Patent Document 1. The potential accident responsibility value can be determined using a map that determines the potential accident responsibility value from the actions of the vehicle and surrounding moving objects predicted by the action prediction unit 828 and the accident responsibility rule. The potential accident responsibility value may be determined only as to whether or not there is responsibility without expressing the potential accident responsibility as a numerical value, or the potential accident responsibility value may be compared with a threshold value to determine whether or not there is responsibility as a result.

[0135] When the responsibility determination unit 830 determines that the vehicle will not be responsible for a potential accident if the vehicle travels along a candidate route, it instructs the autonomous driving function unit 29 to use the candidate route as the route along which the vehicle will travel. When the driving plan unit 27 outputs only one candidate route, and the responsibility determination unit 830 determines that the vehicle will be responsible for a potential accident if the vehicle travels along the candidate route, it instructs the driving plan unit 27 to recalculate the candidate route.

[0136] In addition, when the driving plan unit 27 outputs a plurality of candidate routes as candidate routes to be driven in the same time period, the responsibility determination unit 830 determines the responsibility of the vehicle for a potential accident for each of the plurality of candidate routes. When it is determined that the vehicle is responsible for a potential accident for all of the plurality of candidate routes, the driving plan unit 27 is instructed to recalculate the candidate routes. When it is determined that the vehicle is not responsible for a potential accident for two or more of the plurality of candidate routes, one candidate route is selected based on a criterion determined in advance. Then, the selected candidate route is output to the automatic driving function unit 29. The criterion set in advance may be various criteria, such as a criterion that the route is most consistent with a mid- to long-term driving plan, a criterion that the speed change is the smallest, and the like. This criterion may also be considered as a criterion for determining the priority order of the plurality of candidate routes.

[0137] When the responsibility determination unit 830 determines that the vehicle is responsible for a potential accident, it may simply prohibit driving along the candidate route to be determined, instead of instructing the driving plan unit 27 to recalculate the candidate route. When driving along the candidate route to be determined is prohibited, the driving plan unit 27 may select whether to recalculate the candidate route or to switch to a lower level of manual driving without recalculating the candidate route in response to the prohibition of the candidate route.

[0138] Storage medium 840 is writable and non-volatile, and can output the stored contents to an external device connected to automatic driving device 802. In addition, storage medium 840 itself may be removable from automatic driving device 802 and can be attached to an external device. Storage medium 840 is, for example, a flash memory.

[0139] The storage medium 840 stores the behavior of the prediction target moving object predicted by the behavior prediction unit 828. By storing the behavior of the prediction target moving object predicted by the behavior prediction unit 828 in the storage medium 840, it is possible to verify later whether the behavior prediction unit 828 has correctly predicted the behavior of the prediction target moving object.

[0140] It is preferable that the storage medium 840 stores the behavior of the prediction target moving object predicted by the behavior prediction unit 828 as well as the candidate routes calculated by the travel planning unit 27. In this way, the validity of the behavior of the prediction target moving object predicted by the behavior prediction unit 828 can be verified in consideration of the candidate routes.

[0141] Furthermore, it is preferable that the storage medium 840 stores one or more of the following. One is a time associated with the behavior or candidate route of the prediction target moving object predicted by the behavior prediction unit 828. The other is a judgment result of the responsibility determination unit 830. In addition, one or more of the behavior recognized by the driving environment recognition unit 824, the detection value of the behavior detection sensor 805, and the accident responsibility rule acquired by the scene determination unit 829 may be stored in the storage medium 840.

[0142] <Summary of embodiment 8> The vehicle system 801 of the eighth embodiment includes a responsibility determination unit 830 that determines responsibility for a potential accident that is expected when the vehicle travels along a candidate route. Therefore, if the vehicle travels along a candidate route and an accident occurs, it can be determined in advance whether the vehicle bears responsibility for the accident. If the automatic driving device 802 drives the vehicle along the candidate route using automatic driving control at a level that specifies that the automatic driving device 802 is responsible for the accident, it can be determined in advance whether the automatic driving device 802 bears responsibility for the accident.

[0143] In addition, the behavior prediction unit 828 included in the automatic driving device 802 acquires a potential occurrence probability, which is a potential occurrence probability of the vehicle approaching an obstacle, which can be calculated from environmental information, and predicts the behavior of the prediction target moving object by taking this potential occurrence probability into consideration. This improves the reliability of the behavior prediction of the prediction target moving object. Furthermore, by improving the reliability of the behavior prediction of the prediction target moving object, the reliability of the judgment by the responsibility judgment unit 830 is also improved.

[0144] (Embodiment 9) Next, an eighth embodiment will be described. Fig. 9 shows a configuration diagram of a vehicle system 901 of the ninth embodiment. In the vehicle system 901 of the ninth embodiment, the configuration of an automatic driving unit 926 in the configuration of an automatic driving device 902 is different from the automatic driving unit 826 of the eighth embodiment.

[0145] The automatic driving unit 926 includes a driving plan unit 27, like the automatic driving unit 826 of the eighth embodiment. The automatic driving unit 926 also includes a behavior prediction unit 928, a scene determination unit 929, and a responsibility determination unit 930.

[0146] The driving plan unit 27 calculates a candidate route in the same manner as the driving plan unit 27 in the eighth embodiment. However, while the candidate route is input to the behavior prediction unit 828 and the scene determination unit 829 in the eighth embodiment, the candidate route is not input to the behavior prediction unit 928 or the scene determination unit 929 in the ninth embodiment, but is input to the responsibility determination unit 930.

[0147] The behavior prediction unit 928 acquires the driving environment recognized by the driving environment recognition unit 824, but predicts the behavior of the prediction target moving object without acquiring the candidate route. Therefore, the behavior of the prediction target moving object is predicted from the current behavior of the own vehicle, surrounding vehicles, and other moving objects. For example, for the own vehicle and surrounding vehicles, the future speed is predicted based on the current speed and acceleration, and the future traveling direction is predicted assuming that the current steering angle is maintained. In addition to these, the altitude of the prediction target moving object is predicted taking into consideration various driving environments recognized by the driving environment recognition unit 824. Examples of behavior predicted in consideration of environmental information include, for example, prediction that the vehicle will stop or decelerate due to the light color of a traffic light, prediction that the own vehicle will decelerate due to the presence of a vehicle ahead, and change of traveling direction along the road because the road is curved. Note that the time for the behavior prediction unit 928 to predict the behavior of the prediction target moving object is the same as that of the behavior prediction unit 828 in the eighth embodiment.

[0148] The scene determination unit 929 also acquires the driving environment recognized by the driving environment recognition unit 824, but does not acquire the candidate route. The scene determination unit 929 acquires the current vehicle position and the accident responsibility rule around the current vehicle position. The surroundings of the vehicle position are a range including the range in which the prediction target moving object can act, based on the behavior of the prediction target moving object predicted by the behavior prediction unit 928.

[0149] The responsibility determination unit 930 acquires the candidate route calculated by the travel plan unit 27, the behavior of the predicted target moving object predicted by the behavior prediction unit 928, and the accident responsibility rule acquired by the scene determination unit 929. Then, the responsibility determination unit 930 refers to the behavior of the predicted target moving object predicted by the behavior prediction unit 928, and determines the behavior of the predicted target moving object including the own vehicle when the own vehicle travels along the candidate route. Furthermore, the responsibility determination unit 930 judges the responsibility of the own vehicle for a potential accident that is expected when the own vehicle travels along the candidate route by applying the determined behavior to the accident responsibility rule acquired from the scene determination unit 929.

[0150] <Summary of the 9th embodiment> In this embodiment 9, the responsibility determination unit 930 acquires candidate routes, while the behavior prediction unit 928 predicts the behavior of the prediction target moving object without acquiring candidate routes. In this manner, as with the responsibility determination unit 830 in embodiment 8, it is possible to determine the responsibility of the own vehicle for a potential accident that is assumed when the own vehicle travels along the candidate route.

[0151] In addition, with the configuration of embodiment 9, if an accident occurs and it becomes necessary to verify the accident, it becomes easier to verify whether there was a problem with the candidate route or whether the behavior predicted by the behavior prediction unit 928 was inaccurate.

[0152] (Embodiment 10) In the first, eighth, and ninth embodiments, the automatic driving device 2 and the vehicle control ECU 6 are separate entities, but this is not necessarily limited to the above. For example, the automatic driving device 2 may also assume the functions of the vehicle control ECU 6. The automatic driving device 2 may also assume the functions of the locator 3.

[0153] The present disclosure is not limited to the above-described embodiment, and various modifications are possible within the scope of the claims. The technical scope of the present disclosure also includes embodiments obtained by appropriately combining the technical means disclosed in different embodiments. The control unit and the method described in the present disclosure may be realized by a dedicated computer constituting a processor programmed to execute one or more functions embodied by a computer program. Alternatively, the device and the method described in the present disclosure may be realized by a dedicated hardware logic circuit. Alternatively, the device and the method described in the present disclosure may be realized by one or more dedicated computers configured by a combination of a processor that executes a computer program and one or more hardware logic circuits. The computer program may be stored in a computer-readable non-transient tangible recording medium as instructions executed by the computer.

Claims

1. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability that is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit, The safety distance setting unit sets a safety distance at least ahead of the vehicle, the safety distance adjustment unit increases the safety distance ahead of the vehicle set by the safety distance setting unit in response to an increase in the potential occurrence probability; The safety distance setting unit also estimates a forward safety distance for a following vehicle of the vehicle as the obstacle, The safety distance adjustment unit is an automatic driving device that, when the following vehicle distance, which is the distance between the vehicle and the vehicle following the vehicle as an obstacle, becomes less than the safety distance ahead of the following vehicle estimated by the safety distance setting unit, increases the distance by which the following vehicle distance falls short of the safety distance by adding it to the safety distance ahead of the vehicle set by the safety distance setting unit.

2. The automatic driving device according to claim 1 , wherein the environmental information includes information regarding the frequency with which moving objects appear on a road.

3. A driving planner (27) that calculates one or more routes along which the vehicle will travel by the automatic driving; a confirmation unit (28) that evaluates whether or not the route calculated by the travel planning unit is to be used for autonomous driving based on whether or not a distance between the vehicle and an obstacle around the vehicle is equal to or greater than the safety distance set by the safety distance setting unit; The automatic driving device according to claim 1 or 2, further comprising an automatic driving function unit (29) that performs automatic driving along the route evaluated by the confirmation unit to be used for automatic driving.

4. A recognition unit (24) for recognizing a moving object among the obstacles, The recognition unit also distinguishes and recognizes the type of the moving object, The automatic driving device according to any one of claims 1 to 3, wherein the safety distance adjustment unit changes an increase rate of the safety distance depending on the type of the moving body recognized by the recognition unit when increasing the safety distance set by the safety distance setting unit.

5. The vehicle is provided with a recognition unit (24) that recognizes a moving object among the obstacles, The recognition unit also recognizes the behavior of the moving object, The automatic driving device according to any one of claims 1 to 4, wherein the safety distance adjustment unit changes the increase rate of the safety distance when increasing the safety distance set by the safety distance setting unit, depending on the behavior of the moving body recognized by the recognition unit.

6. The safety distance setting unit sets at least a left and right safety distance of the vehicle, a counting unit (241) that counts the number of moving objects detected around the vehicle while distinguishing between moving objects on the left side of the vehicle and moving objects on the right side of the vehicle, An automatic driving device as described in any one of claims 1 to 5, wherein the safety distance adjustment unit increases only the safety distance on the side where the number of moving bodies counted by the counting unit is greater, of the safety distances on the left and right sides of the vehicle set by the safety distance setting unit, when the potential occurrence probability becomes equal to or greater than a threshold value.

7. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability that is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit, The safety distance setting unit sets at least a left and right safety distance of the vehicle, a counting unit (241) that counts the number of moving objects detected around the vehicle while distinguishing between moving objects on the left side of the vehicle and moving objects on the right side of the vehicle, The safety distance adjustment unit is an automatic driving device that increases only the safety distance on the side with a larger number of moving bodies counted by the counting unit, of the safety distances on the left and right sides of the vehicle set by the safety distance setting unit, when the potential occurrence probability becomes equal to or greater than a threshold value.

8. the safety distance setting unit sets at least a left and right safety distance of the vehicle, and sets the ratio of the left and right safety distances to an equal value as a standard; The automatic driving device of claim 6 or 7, wherein the safety distance adjustment unit changes the ratio of the left and right safety distances of the vehicle by increasing only the safety distance on the side where the number of moving bodies counted by the counting unit is greater, out of the left and right safety distances of the vehicle set by the safety distance setting unit, when the potential occurrence probability becomes equal to or greater than a threshold value.

9. An automatic driving device as described in any one of claims 6 to 8, wherein the safety distance adjustment unit increases only the safety distance on the left and right side of the vehicle set by the safety distance setting unit, which is on the side where the number of moving bodies counted by the counting unit is greater, in accordance with an increase in the ratio of the number of moving bodies counted by the counting unit on the left and right sides of the vehicle.

10. The moving objects detected around the vehicle are also detected by distinguishing types, The automatic driving device according to claim 9 , wherein the counting unit counts the number of the moving bodies by weighting the type of the moving body with a higher priority for protection in terms of traffic.

11. a probability calculation unit (25) that calculates the potential occurrence probability by using environmental information acquired by the environmental information acquisition unit; The automatic driving device according to any one of claims 1 to 10, wherein, when the environmental information used to calculate the potential occurrence probability includes information of a location that is a factor that increases the potential occurrence probability, the probability calculation unit calculates the potential occurrence probability to be lower as the distance between the location and the vehicle increases.

12. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability, which is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit; a probability calculation unit (25) that calculates the potential occurrence probability by using the environmental information acquired by the environmental information acquisition unit, a probability calculation unit (25) that calculates the potential occurrence probability by using environmental information acquired by the environmental information acquisition unit; The probability calculation unit is an automatic driving device that, when the environmental information used to calculate the potential occurrence probability includes information about a location that is a factor that increases the potential occurrence probability, calculates the potential occurrence probability to be lower as the distance between the location and the vehicle increases.

13. The automatic driving device according to any one of claims 1 to 12, wherein the safety distance adjustment unit increases the safety distance set by the safety distance setting unit so that the safety distance is greater than the shortest distance at which the vehicle can be stopped by the automatic driving and is less than the stopping distance that is the sum of the estimated free-running distance and braking distance when the vehicle is stopped by manual driving.

14. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability that is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit, The safety distance adjustment unit is an automatic driving device that increases the safety distance set by the safety distance setting unit so that the safety distance is greater than the shortest distance at which the vehicle can be stopped by automatic driving, and less than the stopping distance that is the sum of the estimated free-running distance and braking distance if the vehicle is stopped by manual driving.

15. A response time adjustment unit (291) capable of adjusting a response time from an instruction to operate a braking device of the vehicle to a start of the operation when the vehicle is stopped by the automatic driving, The automatic driving device according to any one of claims 1 to 14, wherein the response time adjustment unit provides a margin in the response time when the potential occurrence probability is less than a threshold value, while adjusting the response time to be shorter when the potential occurrence probability is equal to or greater than the threshold value than when the potential occurrence probability is less than the threshold value.

16. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability, which is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit; A response time adjustment unit (291) capable of adjusting a response time from an instruction to operate a braking device of the vehicle to a start of the operation when the vehicle is stopped by the automatic driving, The response time adjustment unit provides a margin in the response time when the potential occurrence probability is less than a threshold value, while adjusting the response time to be shorter when the potential occurrence probability is equal to or greater than the threshold value, compared to when the potential occurrence probability is less than the threshold value.

17. A lane changing unit (292) that causes the vehicle to change lanes by the automatic driving, The automatic driving device according to any one of claims 1 to 16, wherein the lane changing unit causes the vehicle to change lanes to a lane with a lower potential occurrence probability when the potential occurrence probability is equal to or greater than a threshold value, when the road on which the vehicle is traveling is a road with multiple lanes in each direction, and when the lane on which the vehicle is traveling is not the lane of the road with multiple lanes in each direction in which the potential occurrence probability is lowest.

18. Used in autonomous vehicles, an environmental information acquisition unit (23) that acquires environmental information, which is information about the environment in which the vehicle is placed other than obstacles around the vehicle; a safety distance setting unit (281) for setting a minimum safety distance from the vehicle to the obstacle in order to avoid the vehicle coming close to the obstacle, using at least information on the behavior of the vehicle; a safety distance adjustment unit (282) that increases the safety distance set by the safety distance setting unit in response to an increase in a potential occurrence probability, which is a potential occurrence probability of close proximity between the vehicle and the obstacle based on the environmental information acquired by the environmental information acquisition unit; A lane changing unit (292) that causes the vehicle to change lanes by the automatic driving, The lane changing unit is an automatic driving device that, when the potential occurrence probability is equal to or greater than a threshold value, when the road on which the vehicle is traveling is a road with multiple lanes in each direction, and when the lane on which the vehicle is traveling is not the lane of the road with multiple lanes in each direction in which the potential occurrence probability is lowest, causes the lane to be changed to a lane with a lower potential occurrence probability.

19. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; The safety distance is set at least in front of the vehicle, Increasing a set safety distance ahead of the vehicle as the potential occurrence probability increases; The safe distance is also estimated ahead for a following vehicle of the vehicle as the obstacle, An autonomous driving method in which, when the following vehicle distance, which is the distance between the vehicle and a vehicle following the vehicle as an obstacle, becomes less than an estimated safe distance ahead of the following vehicle, the following vehicle distance is increased by adding the distance that the following vehicle distance falls short of the safety distance to the set safety distance ahead of the vehicle.

20. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; The safety distance is set at least on the left and right of the vehicle, counting the number of moving objects detected around the vehicle while distinguishing between moving objects on the left side of the vehicle and moving objects on the right side of the vehicle; An autonomous driving method that increases only the safety distance on the left and right sides of the vehicle on the side with a larger number of counted moving objects when the potential occurrence probability is equal to or greater than a threshold value.

21. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; Calculating the potential occurrence probability using the environmental information; An autonomous driving method in which, when the environmental information used to calculate the potential occurrence probability includes information about a location that is a factor in increasing the potential occurrence probability, the potential occurrence probability is calculated to be lower as the distance between the location and the vehicle increases.

22. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; An automatic driving method for increasing the safety distance so that when the safety distance is increased, the safety distance is increased so that it is greater than the shortest distance at which the vehicle can be stopped by automatic driving, and less than the stopping distance which is the sum of the estimated free-running distance and braking distance if the vehicle were to be stopped by manual driving.

23. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; adjusting a response time from an instruction to operate a braking device of the vehicle to a start of the operation when stopping the vehicle by the automatic driving; The autonomous driving method includes adjusting the response time so that, when the potential occurrence probability is less than a threshold value, the response time is provided with a margin of error, while, when the potential occurrence probability is equal to or greater than the threshold value, the response time is adjusted to be shorter than when the potential occurrence probability is less than the threshold value.

24. Used in autonomous vehicles, Acquire environmental information, which is information about the environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle in order to avoid the vehicle coming into close proximity to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; causing the vehicle to change lanes by the automatic driving; An autonomous driving method for changing lanes, in which, when the potential occurrence probability is equal to or greater than a threshold value, the road on which the vehicle is traveling is a road with multiple lanes in each direction, and the lane on which the vehicle is traveling is not the lane of the road with multiple lanes in each direction in which the potential occurrence probability is lowest, the lane is changed to a lane with a lower potential occurrence probability.

25. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; and increasing a minimum safe distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming into close proximity, the minimum safe distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; The safety distance is set at least in front of the vehicle, and increasing a set safety distance ahead of the vehicle as the potential occurrence probability increases; The safe distance is also estimated ahead for a following vehicle of the vehicle as the obstacle, The control program further executes, when the following vehicle distance, which is the distance between the vehicle and a vehicle following the vehicle as the obstacle, becomes less than the estimated safe distance ahead of the following vehicle, increasing the following vehicle distance by the distance that is less than the safety distance, by adding it to the set safety distance ahead of the vehicle.

26. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; and increasing a minimum safe distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming into close proximity, the minimum safe distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; The safety distance is set at least on the left and right of the vehicle, Counting the number of moving objects detected around the vehicle while distinguishing between moving objects on the left side of the vehicle and moving objects on the right side of the vehicle; The control program further executes, when the potential occurrence probability is equal to or greater than a threshold value, increasing only the safety distance on the left and right sides of the vehicle on the side where the number of counted moving bodies is greater.

27. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; and increasing a minimum safe distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming into close proximity, the minimum safe distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; The potential occurrence probability is calculated using the environmental information; The control program further executes, when the environmental information used to calculate the potential occurrence probability includes information on a location that is a factor in increasing the potential occurrence probability, the control program calculates the potential occurrence probability to be lower as the distance between the location and the vehicle increases.

28. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; and increasing a minimum safe distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming into close proximity, the minimum safe distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming into close proximity to the obstacle based on the acquired environmental information; A control program that executes the following to increase the safety distance so that, when the safety distance is increased, the safety distance is increased to a range that is greater than the shortest distance at which the vehicle can be stopped by the automatic driving and less than the stopping distance that is the sum of the estimated free running distance and braking distance if the vehicle is stopped by manual driving.

29. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming close to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming close to the obstacle based on the acquired environmental information; adjusting a response time from an instruction to operate a braking device of the vehicle to a start of the operation when the vehicle is stopped by the automatic driving; The adjustment of the response time includes providing a margin in the response time when the potential occurrence probability is less than a threshold value, and adjusting the response time to be shorter when the potential occurrence probability is equal to or greater than the threshold value than when the potential occurrence probability is less than the threshold value.

30. A control program for use in an autonomous vehicle, The processor: acquiring environmental information that is information about an environment in which the vehicle is located other than obstacles around the vehicle; increasing a minimum safety distance that the vehicle should keep between itself and the obstacle to avoid the vehicle coming close to the obstacle, the minimum safety distance being set using at least information on the behavior of the vehicle, in accordance with an increase in a potential occurrence probability that is a potential occurrence probability of the vehicle coming close to the obstacle based on the acquired environmental information; causing the vehicle to change lanes by the automatic driving; A control program that executes, when the potential occurrence probability becomes equal to or exceeds a threshold value during lane change, when the road on which the vehicle is traveling is a road with multiple lanes in each direction, and when the lane on which the vehicle is traveling is not the lane of the road with multiple lanes in each direction in which the potential occurrence probability is lowest, the lane change is made to a lane with a lower potential occurrence probability.

Images