Driving assistance device, driving assistance processing method, and recording medium
The driving support device dynamically adjusts vehicle trajectory and speed based on traffic rule strength and time-dependent changes, addressing the limitations of current technologies by ensuring appropriate adherence to traffic rules and enhancing safety and reliability.
Patent Information
- Application Number
- PCT/JP2023/019063
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-05-23
- Publication Date
- 2024-11-28
AI Technical Summary
Current driving support technologies fail to adequately reflect the strength of traffic rules and changes in traffic rules over time, leading to inadequate driving support control, as they do not account for varying tolerances in boundary lines, traffic control zones, and time-dependent changes such as traffic light colors or bus lanes.
A driving support device and method that acquires information on the vehicle's surroundings to set a 'rule deviation risk' based on traffic rule strength and time, allowing for dynamic adjustment of the vehicle's trajectory and speed to maintain compliance with traffic rules, with risks increasing proportionally with vehicle speed.
This approach enables appropriate driving support control by reflecting the strength of traffic rules and temporal changes, enhancing safety and reliability by setting driving conditions that align with individual driver perceptions and traffic rule adherence.
Smart Images

Figure JP2023019063_28112024_PF_FP_ABST
Abstract
Description
Driving assistance device, driving assistance processing method, and recording medium
[0001] The present disclosure relates to a driving assistance device, a driving assistance processing method, and a recording medium.
[0002] In recent years, research and development of autonomous driving technology and driving assistance technology has been progressing with the aim of preventing and reducing accidents and reducing driving burden. For example, Patent Literature 1 discloses a method for autonomous driving technology and driving assistance technology that uses the concept of risk potential as one of the algorithms for setting a vehicle's driving trajectory and speed. The risk potential represents the magnitude of the risk of a vehicle colliding with an obstacle on the trajectory or the risk of the vehicle running off the road, with the risk value increasing the closer the vehicle is to an obstacle or road edge. A control device sets a trajectory and speed with low risk based on a risk map generated by superimposing risk potentials set for multiple obstacles and road edges.
[0003] For example, when passing by a parked vehicle on the road, some drivers will slow down sufficiently to pass by, while others will not slow down at all, meaning that the subjective risk for each driver will differ even in the same situation. Non-Patent Document 1 proposes a technology that uses the concept of acceptable risk, which differs for each driver. The acceptable risk is a concept related to the vehicle speed, and is used to reduce the speed when the risk value in the traveling direction exceeds the acceptable risk, thereby creating a region where the risk value is below the acceptable risk.
[0004] Japanese Patent Application Laid-Open No. 2022-083196
[0005] A Study on the Application of a Driver Model Based on the Perception of Danger in the Forward Vision to Autonomous Driving Control (Transactions of the Society of Automotive Engineers of Japan, Vol. 48, No. 3, pp. 763-769)
[0006] However, the technology disclosed in Non-Patent Document 1 does not reflect the strength of traffic rules or responses to changes in traffic rules over time. Specifically, for example, there are multiple types of boundary lines for vehicle lane boundaries, each with different tolerance levels for vehicle deviation from the boundary line (traffic rule strength). Furthermore, roads are not limited to boundary lines, but also have traffic restriction zones, and the tolerance levels for entry into traffic restriction zones (traffic rule strength) vary depending on the type of traffic restriction zone. Furthermore, while boundary lines and traffic restriction zones are mostly set on roads and are unchanging, there are also cases where traffic rules change over time even within the same driving area, such as traffic light colors and bus lanes.
[0007] In the technology disclosed in Non-Patent Document 1, there is only one acceptable risk for each driver, and it is not possible to set driving conditions that reflect the strength of traffic rules or changes in traffic rules over time.
[0008] The present invention has been made in consideration of the above problems, and an object of the present invention is to provide a driving assistance device, a driving assistance processing method, and a storage medium that are capable of executing appropriate driving assistance control by reflecting the strength of traffic rules and changes in traffic rules over time.
[0009] In order to solve the above problem, according to one aspect of the present invention, there is provided a driving assistance device that assists in driving a vehicle, comprising one or more processors and one or more memories communicatively connected to the one or more processors, wherein the one or more processors perform the following operations: an acquisition process that acquires information about the environment surrounding the vehicle; a risk setting process that sets a rule deviation risk, which indicates the risk of deviating from a traffic rule related to the driving of the vehicle, for the target of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule applies; an allowable risk setting process that sets an allowable rule deviation risk, which indicates the allowable range for driving of the vehicle relative to the rule deviation risk, so that it increases in proportion to the vehicle speed of the vehicle; and a driving condition setting process that sets a target trajectory and target speed for the vehicle during automatic driving based on the rule deviation risk and the allowable rule deviation risk so that the vehicle drives at a position where the allowable rule deviation risk exceeds the rule deviation risk.
[0010] In addition, in order to solve the above problem, according to another aspect of the present invention, a driving assistance processing method is provided in which a computer acquires information about the environment surrounding a vehicle, sets a rule deviation risk, which indicates the risk of deviating from a traffic rule related to the driving of the vehicle, for the target of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule applies, sets an acceptable rule deviation risk, which indicates the acceptable range of driving of the vehicle relative to the rule deviation risk, to increase in proportion to the speed of the vehicle, and sets a target trajectory and target speed for the vehicle during automatic driving based on the rule deviation risk and the acceptable rule deviation risk so that the vehicle drives at a position where the acceptable rule deviation risk exceeds the rule deviation risk.
[0011] In addition, in order to solve the above problem, according to another aspect of the present invention, there is provided a non-transitory tangible recording medium having recorded thereon a computer program that causes a computer to perform the following operations: acquire information about the environment surrounding a vehicle; set a rule deviation risk, which indicates the risk of deviating from a traffic rule related to the driving of the vehicle, for the target of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule applies; set an acceptable rule deviation risk, which indicates the acceptable range of driving of the vehicle relative to the rule deviation risk, to increase in proportion to the speed of the vehicle; and set a target trajectory and target speed for the vehicle during automatic driving based on the rule deviation risk and the acceptable rule deviation risk so that the vehicle drives at a position where the acceptable rule deviation risk exceeds the rule deviation risk.
[0012] As described above, according to the present invention, it is possible to execute appropriate driving assistance control by reflecting the strength of traffic rules and changes over time in traffic rules.
[0013] 10 is a schematic diagram showing an example of the configuration of a vehicle equipped with a driving assistance device according to a first embodiment of the present disclosure. FIG. 11 is a block diagram showing an example of the configuration of a driving assistance device according to the present embodiment. FIG. 12 is an explanatory diagram showing a risk potential set for an object. FIG. 13 is an explanatory diagram showing an example in which the tolerable risk is increased by reducing the vehicle speed. FIG. 14 is an explanatory diagram showing an example of a method for setting a driving trajectory and a vehicle speed using a basic method that reflects driver characteristics. FIG. 15 is an explanatory diagram showing an example of a method for setting a driving trajectory and a vehicle speed using a basic method that reflects driver characteristics. FIG. 16 is an explanatory diagram showing an example of a scene to which the driving assistance processing according to the present embodiment can be applied. FIG. 17 is an explanatory diagram showing the risk potential in the scene shown in FIG. 8 as viewed in the direction of travel of the vehicle. FIG. 18 is an explanatory diagram showing a collision risk as viewed in the direction of travel of the vehicle. FIG. 19 is an explanatory diagram showing a rule deviation risk as viewed in the direction of travel of the vehicle. FIG. 19 is an explanatory diagram showing an example in which the strength of traffic rules is set for each type of boundary line and traffic regulation zone. FIG. 19 is an explanatory diagram showing an example in which the rule deviation risk differs depending on the type of boundary line. 10 is an explanatory diagram showing a zebra zone, which is a guidance zone just before the point where the vehicle lane widens to two lanes. FIG. 11 is an explanatory diagram showing a rule deviation risk set for a zebra zone. FIG. 12 is an explanatory diagram showing an acceptable collision risk set for a collision risk. FIG. 13 is an explanatory diagram showing an acceptable rule deviation risk set for a rule deviation risk. FIG. 14 is an explanatory diagram showing an acceptable road departure risk set for a road departure risk. FIG. 15 is an explanatory diagram showing an acceptable collision risk and an acceptable rule deviation risk set when a vehicle is traveling at a relatively low speed. FIG. 16 is an explanatory diagram showing an acceptable collision risk and an acceptable rule deviation risk set when a vehicle is traveling at a relatively high speed. FIG. 17 is an explanatory diagram showing an acceptable driving range set for a driver who feels great resistance to straying into an oncoming lane in the scene shown in FIG. 8. FIG. 18 is an explanatory diagram showing an acceptable driving range set for a driver who feels great danger of hitting a parked vehicle in the scene shown in FIG. 8. FIG. 19 is an explanatory diagram showing an example in which an acceptable driving range is not formed in which the acceptable collision risk exceeds the collision risk and the acceptable rule deviation risk exceeds the rule deviation risk.1 is an explanatory diagram showing an example in which a driving trajectory is set on the side of an oncoming lane by increasing the set vehicle speed in order to create an allowable driving tolerance range. FIG. 1 is a flowchart showing a learning process by the driving assistance device according to the present embodiment. FIG. 2 is a flowchart showing a driving assistance process by the driving assistance device according to the present embodiment. FIG. 3 is a flowchart showing a process by the driving assistance device according to the present embodiment for setting a risk potential and an acceptable risk for a risk object on a road by the driving assistance device according to the present embodiment. FIG. 4 is a flowchart showing a process by the driving assistance device according to the present embodiment for setting a risk potential and an acceptable risk for an obstacle by the driving assistance device according to the present embodiment. FIG. 5 is a flowchart showing a process by the driving assistance device according to the present embodiment for setting a target trajectory and a target speed. FIG. 6 is an explanatory diagram showing a situation in which an example of a second embodiment of the present disclosure can be applied. FIG. 7 is a flowchart showing a process by which a risk potential and an acceptable risk are set according to the example of the present embodiment. FIG. 7 is an explanatory diagram showing a situation in which an example of a third embodiment of the present disclosure can be applied. FIG. 8 is an explanatory diagram showing an example in which the intensity of a traffic rule that changes depending on the lighting color of a traffic signal is set. FIG. 9 is an explanatory diagram showing a risk map when the lighting color of a traffic signal is green. FIG. 10 is an explanatory diagram showing a risk map when the lighting color of a traffic signal is yellow. FIG. 11 is an explanatory diagram showing a risk map when the lighting color of a traffic signal is red. 1 is an explanatory diagram showing a risk map set in a state two minutes before the start of a time period in which a bus-only lane is in use, and FIG. 2 is an explanatory diagram showing a risk map set in a state at the start of a time period in which a bus-only lane is in use.
[0014] Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In this specification and drawings, components having substantially the same functional configurations are designated by the same reference numerals, and redundant explanations will be omitted.
[0015] <<1. First embodiment>> <1-1. Overall configuration of vehicle> First, an example of the overall configuration of a vehicle to which a driving assistance device according to a first embodiment of the present disclosure can be applied will be described.
[0016] Fig. 1 is a schematic diagram showing an example configuration of a vehicle 1 equipped with a driving assistance device 50. The vehicle 1 shown in Fig. 1 is configured as a two-wheel drive vehicle in which a driving torque output from a driving force source 9 that generates driving torque for the vehicle is transmitted to the left and right front wheels. The driving force source 9 may be an internal combustion engine such as a gasoline engine or a diesel engine, a driving motor, or both an internal combustion engine and a driving motor.
[0017] Vehicle 1 may be a four-wheel drive vehicle that transmits drive torque to the front, rear, left, and right wheels. In this case, vehicle 1 may be an electric vehicle equipped with two drive motors, for example, a front-wheel drive motor and a rear-wheel drive motor, or an electric vehicle equipped with drive motors corresponding to each wheel. Furthermore, if vehicle 1 is an electric vehicle or a hybrid electric vehicle, vehicle 1 is equipped with a power generation device such as a secondary battery that stores power supplied to the drive motors, and a motor or fuel cell that generates power to charge the secondary battery.
[0018] The vehicle 1 is equipped with a driving force source 9, an electric steering device 15, and a brake control unit 16 as devices used to control the operation of the vehicle 1. The driving force source 9 outputs driving torque that is transmitted to the front wheel drive shaft 5F via a transmission and a front wheel differential mechanism 7 (not shown). The operation of the driving force source 9 and the transmission is controlled by a vehicle control device 41 that includes one or more electronic control units (ECUs: Electronic Control Units).
[0019] The front-wheel drive shaft 5F is provided with an electric steering device 15. The electric steering device 15 includes an electric motor and a gear mechanism (not shown), and is controlled by a vehicle control device 41 to adjust the steering angle of the left and right front wheels. During manual driving, the vehicle control device 41 controls the electric steering device 15 based on the steering angle of the steering wheel 13 by the driver. During automatic driving, the vehicle control device 41 controls the electric steering device 15 based on a target steering angle set by the driving assistance device 50.
[0020] The brake control unit 16 controls the hydraulic pressure supplied to brake devices 17LF, 17RF, 17LR, and 17RR (hereinafter collectively referred to as brake devices 17) provided on the front, rear, left, and right wheels, respectively. Each brake device 17 is configured as a device that applies braking force to the wheel by, for example, clamping a brake disc that rotates together with the wheel with brake pads using the supplied hydraulic pressure.
[0021] The brake control unit 16 includes an electric motor pump that discharges brake fluid, and a plurality of solenoid valves that adjust the hydraulic pressure supplied to each brake device 17. The brake control unit 16 is controlled by the vehicle control device 41 to adjust the hydraulic pressure supplied to each brake device 17 and generate a predetermined braking force on each of the front, rear, left, and right wheels. When the driving force source 9 includes a drive motor, the hydraulic braking by the brake control unit 16 is used in combination with regenerative braking using the drive motor.
[0022] The vehicle control device 41 includes one or more electronic control devices that control the driving of the driving force source 9, the electric steering device 15, and the brake control unit 16. The vehicle control device 41 may also have a function of controlling the driving of a transmission that changes the speed of the output from the driving force source 9 and transmits it to the wheels 3. The vehicle control device 41 is configured to be able to acquire information transmitted from the driving assistance device 50, and is configured to be able to execute automatic driving control of the vehicle 1.
[0023] The vehicle 1 also includes an interior camera 21 , front cameras 23LF and 23RF, a vehicle state sensor 25 , a GNSS (Global Navigation Satellite System) sensor 27 , a map data storage unit 29 , and a notification device 43 .
[0024] The interior camera 21 captures images of the interior of the vehicle and generates image data. The interior camera 21 is equipped with an imaging element such as a CCD (Charged-Coupled Device) or a CMOS (Complementary Metal-Oxide-Semiconductor), and outputs the generated image data to the driving assistance device 50. In this embodiment, the interior camera 21 is installed so as to be able to capture images of the driver of the vehicle 1.
[0025] The front imaging cameras 23LF, 23RF are one form of ambient environment sensors for acquiring information about the ambient environment of the vehicle 1. The front imaging cameras 23LF, 23RF capture images of the area ahead of the vehicle 1 and generate image data. The front imaging cameras 23LF, 23RF are equipped with imaging elements such as CCD or CMOS, and output the generated image data to the driving assistance device 50.
[0026] 1, the front imaging cameras 23LF, 23RF are configured as a pair of left and right stereo cameras, but may also be monocular cameras. The vehicle 1 may also be provided with, for example, a rear imaging camera in addition to the front imaging cameras 23LF, 23RF. Furthermore, the vehicle 1 may also be provided with, as a surrounding environment sensor, one or more sensors other than cameras, such as LiDAR (Light Detection and Ranging), millimeter-wave radar, or ultrasonic sensors, in addition to the front imaging cameras 23LF, 23RF.
[0027] The vehicle state sensor 25 is composed of at least one sensor that detects the operation state and behavior of the vehicle 1. The vehicle state sensor 25 includes at least one of a steering angle sensor, an accelerator position sensor, a brake stroke sensor, a brake pressure sensor, or an engine rotation speed sensor, and detects the operation state of the vehicle 1, such as the steering angle of the steering wheel or steered wheels, the accelerator opening, the amount of brake operation, or the engine rotation speed. The vehicle state sensor 25 also includes at least one of a vehicle speed sensor, an acceleration sensor, or an angular velocity sensor, and detects the behavior of the vehicle, such as the vehicle speed, longitudinal acceleration, lateral acceleration, and yaw rate. The vehicle state sensor 25 outputs a sensor signal including the detected information to the driving assistance device 50.
[0028] The GNSS sensor 27 receives satellite signals transmitted from satellites of a positioning system such as the Global Positioning System (GPS). The GNSS sensor 27 outputs position information of the vehicle 1 contained in the received satellite signals to the driving assistance device 50. The position information of the vehicle 1 is acquired as, for example, latitude and longitude information. Note that the GNSS sensor 27 may be a sensor other than a GPS sensor that receives satellite signals from satellites of other positioning systems that identify the position of the vehicle 1.
[0029] The map data storage unit 29 is configured with a storage medium such as a hard disk drive (HDD), a compact disc (CD), a digital versatile disc (DVD), a solid state drive (SSD), a universal serial bus (USB) flash drive, or a storage device, and stores map data. The stored map data includes data on reference paths, which are trajectories that serve as a reference for traveling on each road. The map data storage unit 29 may be a storage medium that stores map data of a navigation system.
[0030] The notification device 43 is driven by the driving assistance device 50 and presents various information to the driver by means of image display, audio output, etc. The notification device 43 includes, for example, a display device provided in the instrument panel and a speaker provided in the vehicle 1. The display device may be a display device of a navigation system. The notification device 43 may also include a HUD (head-up display) that displays information on the front window superimposed on the scenery around the vehicle 1.
[0031] The driving assistance device 50 executes control to assist in driving the vehicle 1. In this embodiment, the driving assistance device 50 causes the vehicle 1 to travel in an autonomous driving manner based on information about the surrounding environment output from surrounding environment sensors including the forward-facing imaging cameras 23LF, 23RF, data about the vehicle's operation state and behavior transmitted from the vehicle state sensor 25, and position information about the vehicle 1 transmitted from the GNSS sensor 27. Specifically, the driving assistance device 50 acquires reference path data stored in the map data storage unit 29, sets driving conditions for the vehicle 1 so that the vehicle 1 does not come into contact with pedestrians or obstacles (hereinafter, pedestrians and obstacles will be collectively referred to as "obstacles" unless otherwise noted), and transmits a control command to the vehicle control device 41.
[0032] <1-2. Driving Assistance Device> Next, the driving assistance device 50 according to the first embodiment of the present disclosure will be specifically described.
[0033] The driving assistance device 50 of this embodiment is configured to execute an acquisition process to acquire information about the surrounding environment of the vehicle 1, a risk setting process to set a rule deviation risk, which indicates the risk of deviating from traffic rules related to the driving of the vehicle 1, for the target of the traffic rule in accordance with the strength of the traffic rule, an acceptable risk setting process to set an acceptable rule deviation risk, which indicates the acceptable range of driving of the vehicle 1 relative to the rule deviation risk, so that it increases in proportion to the vehicle speed of the vehicle 1, and a driving condition setting process to set a target trajectory and target speed for the automatic driving of the vehicle 1 based on the rule deviation risk and the acceptable rule deviation risk so that the vehicle 1 drives at a position where the acceptable rule deviation risk exceeds the rule deviation risk.
[0034] 2 is a block diagram showing an example of the configuration of the driving assistance device 50 according to this embodiment. The driving assistance device 50 functions as a device that recognizes objects by having one or more processors, such as a central processing unit (CPU) or a graphics control unit (GPU), execute a computer program. The computer program is a computer program that causes the processor to execute operations, described below, that should be performed by the driving assistance device 50. The computer program executed by the processor may be recorded on a recording medium that functions as a storage unit (memory) provided in the driving assistance device 50, or may be recorded on a recording medium built into the driving assistance device 50 or any recording medium that can be externally attached to the driving assistance device 50.
[0035] The recording medium for recording a computer program may be a magnetic medium such as a hard disk, a floppy disk, or a magnetic tape; an optical recording medium such as a CD-ROM, a DVD, or a Blu-ray (registered trademark); a magneto-optical medium such as a floptical disk; a memory element such as a RAM or a ROM; a flash memory such as a USB memory or an SSD; or any other medium capable of storing a program.
[0036] Note that part or all of the driving assistance device 50 may be configured as updatable firmware or the like, or may be a program module or the like that is executed by instructions from a CPU or the like.
[0037] The driving assistance device 50 is connected to an in-vehicle camera 21, a surrounding environment sensor 23, a vehicle state sensor 25, a GNSS sensor 27, and a map data storage unit 29 via a dedicated line or communication means such as a controller area network (CAN) or local internet (LIN). A vehicle control device 41 and a notification device 43 are also connected to the driving assistance device 50. Note that the driving assistance device 50 is not limited to an electronic control device mounted on the vehicle 1, and may be a terminal device such as a smartphone or a wearable device. As shown in FIG. 2 , the driving assistance device 50 includes a processing unit 51, a storage unit 53, and a learning data storage unit 55.
[0038] (1-2-1. Storage Unit) The storage unit 53 is configured by a storage element such as a RAM (Random Access Memory) or a ROM (Read Only Memory), or a storage medium such as an HDD, CD, DVD, SSD, USB flash, storage device, etc. The storage unit 53 stores software programs executed by the arithmetic processing device, various parameters used in arithmetic processing, acquired data, arithmetic results, etc.
[0039] (1-2-2. Learning Data Storage Unit) The learning data storage unit 55 is configured with an updatable storage medium such as RAM, HDD, CD, DVD, SSD, USB flash, or storage device, and stores the learning data learned by the processing unit 51. In this embodiment, the learning data includes the following three acceptable risks: - Acceptable collision risk indicating the acceptable driving range for the risk perceived by the driver in relation to obstacles around the vehicle (hereinafter also referred to as "collision risk"); - Acceptable rule deviation risk indicating the acceptable driving range for the risk perceived by the driver in relation to deviation from traffic rules (hereinafter also referred to as "rule deviation risk"); - Acceptable road departure risk indicating the acceptable driving range for the risk perceived by the driver in relation to running off the road edge (hereinafter also referred to as "road departure risk").
[0040] The allowable collision risk, the allowable rule deviation risk, and the allowable road departure risk are each given as a formula in which the allowable risk changes depending on the vehicle speed.The allowable collision risk, the allowable rule deviation risk, and the allowable road departure risk will be explained in detail later.
[0041] (1-2-2. Processing Unit) The processing unit 51 is configured to include one or more arithmetic processing devices (processors). The processing unit 51 performs various arithmetic processing by executing programs stored in the memory unit 53. In this embodiment, the processing unit 51 includes an acquisition unit 61, a surrounding environment detection unit 63, a learning processing unit 65, a risk setting unit 67, an acceptable risk setting unit 69, a driving condition setting unit 71, and a notification processing unit 73. Some or all of these units are functions realized by the execution of a program by the arithmetic processing device. However, some of the units constituting the processing unit 51 may be configured as hardware circuits such as analog circuits.
[0042] (1-2-2-1. Acquisition unit) The acquisition unit 61 acquires information or data output from the interior camera 21, the surrounding environment sensor 23, the vehicle state sensor 25, and the GNSS sensor 27. The acquisition unit 61 acquires information or data at appropriate time intervals from the interior camera 21, the surrounding environment sensor 23, the vehicle state sensor 25, and the GNSS sensor 27. The acquisition unit 61 stores the acquired information and data in the storage unit 53 as time-series data.
[0043] (1-2-2-2. Surrounding environment detection unit) The surrounding environment detection unit 63 detects the surrounding environment of the vehicle 1 based on information output from the surrounding environment sensor 23. Specifically, the surrounding environment detection unit 63 detects information about obstacles and roads around the vehicle 1 based on image data output from the forward imaging cameras 23LF, 23RF and detection data output from a LiDAR or radar sensor, etc. Obstacles include people, other vehicles, bicycles, buildings, natural objects, and other obstacles that exist around the vehicle 1. The surrounding environment detection unit 63 calculates the positions of obstacles, etc. relative to the vehicle 1, and the distance and relative speed between the vehicle 1 and the obstacles, etc.
[0044] The road information also includes information on roadside boundaries or curbs, lane boundaries, the type of boundary (color or line shape), and restricted driving zones painted on the road, as well as the type of restricted driving zone. The surrounding environment detection unit 63 calculates the distance from the vehicle 1 to the boundary, curb, or restricted driving zone. The distance from the vehicle 1 to the boundary, curb, or restricted driving zone includes the amount of overhang (negative distance) when the vehicle 1 crosses a boundary line, etc. The surrounding environment detection unit 63 stores the detected information on the surrounding environment of the vehicle 1 in the memory unit 53 as time-series data.
[0045] The surrounding environment detection unit 63 may acquire information transmitted from a device outside the vehicle via communication means such as vehicle-to-vehicle communication, road-to-vehicle communication, or a mobile communication network, and identify part of the information about obstacles around the vehicle 1. The surrounding environment detection unit 63 may also identify the position of the vehicle 1 on map data using the position information of the vehicle 1 acquired by the GNSS sensor 27, and identify information about obstacles and the like around the vehicle 1.
[0046] (1-2-2-3. Learning Processing Unit) The learning processing unit 65 learns the acceptable collision risk, the acceptable rule deviation risk, and the acceptable road departure risk for each driver based on various data stored as time-series data in the memory unit 53. In this embodiment, the learning processing unit 65 learns the acceptable collision risk, the acceptable rule deviation risk, and the acceptable road departure risk based on information about the surrounding environment of the vehicle 1 detected by the surrounding environment detection unit 63 and data on the operation state and behavior of the vehicle 1 output from the vehicle state sensor 25.
[0047] Specifically, the learning processing unit 65 extracts information about the environment surrounding the vehicle 1 and data on the operation state and behavior of the vehicle 1 in a situation where the driver of the vehicle 1, while manually driving, corrects the driving trajectory to ensure a sufficient distance from an obstacle. The information about the environment surrounding the vehicle 1 includes, for example, data on the type of obstacle, the size of each obstacle, the distance from the obstacle, and the relative speed from the obstacle. The information about the environment surrounding the vehicle 1 also includes data on the type of road edge, boundary line, or restricted driving zone, and the distance to the road edge, boundary line, or restricted driving zone. The data on the operation state and behavior of the vehicle includes data on the vehicle speed.
[0048] The learning processing unit 65 learns the acceptable collision risk, the acceptable rule deviation risk, and the acceptable road departure risk based on the extracted information on the surrounding environment of the vehicle 1 and the data on the vehicle's operation state and behavior. The learning process for the acceptable collision risk, the acceptable rule deviation risk, and the acceptable road departure risk will be described in detail later.
[0049] (1-2-2-4. Risk Setting Unit) The risk setting unit 67 sets a risk potential for a risk object contained in the surrounding environment detected in the traveling direction of the vehicle 1. In this embodiment, the risk setting unit 67 sets a collision risk indicating the risk of a collision with an obstacle, a rule deviation risk indicating the risk of deviating from traffic rules related to the traveling of the vehicle 1, and a road departure risk indicating the risk of the vehicle 1 running off the edge of the road. The risk setting unit 67 sets the rule deviation risk for a traffic rule object in accordance with the strength of the traffic rule. The risk setting process by the risk setting unit 67 will be described in detail later.
[0050] (1-2-2-5. Acceptable Risk Setting Unit) The acceptable risk setting unit 69 sets an acceptable risk corresponding to each driver stored in the learning data storage unit 55 for each of the collision risk, rule deviation risk, and road departure risk set by the risk setting unit 67. The acceptable risk setting unit 69 sets the acceptable collision risk, which indicates the acceptable range of driving of the vehicle 1 against the collision risk, and the acceptable road departure risk, which indicates the acceptable range of driving of the vehicle 1 against the road departure risk, to decrease in proportion to the speed of the vehicle 1. Furthermore, the acceptable risk setting unit 69 sets the acceptable rule deviation risk, which indicates the acceptable range of driving of the vehicle 1 against the rule deviation risk, to increase in proportion to the speed of the vehicle 1. The details of the acceptable risk setting process by the acceptable risk setting unit 69 will be explained in detail later.
[0051] (1-2-2-6. Driving Condition Setting Unit) The driving condition setting unit 71 sets driving conditions for the vehicle 1 when executing autonomous driving control, and transmits information on the set driving conditions to the vehicle control device 41. The driving condition setting unit 71 sets at least a target trajectory and a target speed so that the vehicle 1 does not come into contact with an obstacle when the vehicle 1 is driven autonomously along a driving route to a set destination, and transmits a control command to the vehicle control device 41. In this embodiment, the driving condition setting unit 71 sets a target trajectory and a target speed for the vehicle 1 using a risk map in which a collision risk, a rule deviation risk, and a road departure risk are set for the environment around the vehicle 1. Details of the driving condition setting process by the driving condition setting unit 71 will be described in detail later.
[0052] (1-2-2-7. Notification Processing Unit) The notification processing unit 73 controls the driving of the notification device 43, thereby controlling notifications to the driver. In this embodiment, after the autonomous driving control of the vehicle 1 is performed in accordance with the driving conditions set based on the risk map, the notification processing unit 73 notifies the driver of the set driving conditions. For example, when the driving trajectory is deviated from the reference path to pass by a detected pedestrian, the notification processing unit 73 notifies the driver that "the vehicle passed on the left side of the road to ensure a safe distance from the pedestrian," or when the vehicle speed is reduced, the notification processing unit 73 notifies the driver that "the vehicle has slowed down to ensure the safety of the pedestrian." The notification is performed, for example, by at least one of audio and visual means.
[0053] <1-3. Example of Operation of Driving Assistance Device> Next, an example of operation of the driving assistance device 50 according to this embodiment will be described.
[0054] (1-3-1. Method for Setting Driving Conditions) First, we will explain the method for setting driving conditions during automated driving by the driving condition setting unit 71. Below, we will explain an example of the basic concept of the method for setting driving conditions, and then we will explain the method for reflecting in the driving conditions the risk levels that each driver tolerates for individual obstacles, going off the road edge, and deviation from traffic rules.
[0055] (Basic Concept) FIG. 3 is an explanatory diagram showing the risk (risk potential) set for a risk object. In FIG. 3, a pedestrian is shown as an example of the risk object. The value of the risk potential (risk value) R i The risk value R increases as the vehicle approaches the risk object (pedestrian), and is at its maximum when the risk object is present. i is the distance from the risk object x i can be expressed as an exponential function of , for example, as shown in the following formula (1).
[0056]
[0057] R i : Risk value C i : Risk absolute value (gain) x i : Distance from the center of the risk object τ i : gradient coefficient r i : Radius of risk object i: Number to distinguish risk object
[0058] Distance from risk object x i -r i The absolute risk value C is the risk value when i is set for each risk object as a value that depends on the risk object. For example, if the risk object is a "pedestrian" or a "low curb", the risk absolute value C for "pedestrian" is set as follows, assuming that a collision with a pedestrian has a higher risk than a collision with a low curb. i is the absolute risk value C for "low curb" i The gradient coefficient τ is set to a value greater than i may be set to a different value for each risk target, or may be set to the same value regardless of the risk target.
[0059] While the vehicle 1 is traveling, a risk potential is assigned to each risk object detected, and the spatial overlap of each risk potential is added up to generate a risk map that takes into account the risk of collision with multiple obstacles. In such a risk map, the level of risk can be shown as contour lines on a two-dimensional plane. Here, the risk value in the risk map is expressed as R RM However, as mentioned above, the risk value RRM has a two-dimensional distribution. This makes it possible to select a trajectory with low risk. At this time, the risk map may be calculated taking into account not only actual risks but also unactual risks (latent risks). For example, when a vehicle turns and passes through a blind spot caused by an obstruction, a potential risk may be added assuming that a passerby may suddenly appear from the blind spot, and this may be reflected in the risk map.
[0060] (Basic Method for Reflecting Driver Characteristics) Here, the risk perceived by each driver is not necessarily the same as physical risk. For example, when driving on a narrow residential road, some drivers feel a strong risk from utility poles jutting into the road. Furthermore, some drivers feel a strong risk from deviating from traffic rules, such as crossing the boundary line of a lane or the boundary line of the road edge, or crossing a restricted driving zone. For this reason, in this embodiment, as a basic method for setting a driving trajectory, the concept of an index of risk level tolerable for each driver (hereinafter also referred to as "tolerable risk") proposed in the publicly known document "Research on the Application of a Driver Model Based on the Sense of Danger in Forward Vision to Autonomous Driving Control" (Transactions of the Society of Automotive Engineers of Japan, Vol. 48, No. 3, pp. 763-769) is used.
[0061] To briefly explain the basic method using acceptable risk, the driver's acceptable risk U a is assumed to be related to the vehicle speed V of the host vehicle, and can be expressed by the following equation (2).
[0062]
[0063] U a : Acceptable risk w a : Gain V: Vehicle speed
[0064] Gain w a The value of the allowable risk U is set for each individual driver. a The risk value R RM is the allowable risk U aIf the vehicle speed V is greater than the allowable risk U a becomes higher, and the risk value R RM is the allowable risk U a It is possible to create an area below this.
[0065] FIG. 4 shows the relationship between the allowable risk U and the vehicle speed V. a 4 is a diagram showing a cross section perpendicular to the vehicle's direction of travel of a risk map that adds up the spatial overlap between the risk potential of the side wall Z1 and the risk potential of the pedestrian Y1 for a vehicle speed V of 30 km / h and a vehicle speed V of 20 km / h.
[0066] In each cross-sectional view and plan view showing the risk map or risk potential in Figure 4 and subsequent figures, the host vehicle is shown as a mass point, ignoring its size, in order to make the explanation easier to understand.
[0067] As shown on the left side of FIG. 4, when the vehicle speed V is 30 km / h, the risk value R RM is the allowable risk U a On the other hand, as shown on the right side of FIG. 4, when the vehicle speed V is 20 km / h, the allowable risk U a The value of increases, and the risk value R RM is the allowable risk U a An area (drivable route) as follows is generated ahead in the traveling direction: By adjusting the vehicle speed V in this way, it becomes possible to set a travel trajectory within the drivable route.
[0068] 5 to 7 are explanatory diagrams showing examples of a method for setting a driving trajectory and a vehicle speed using a basic method that reflects driver characteristics. When traveling along a set driving route to a destination, data on a reference path Pr of the road being traveled is acquired by referring to pre-stored map data. As shown in FIG. 5, when traveling on a road with a side wall Z1 on the left side, the allowable risk U corresponding to the current vehicle speed V (=30 km / h) on the acquired reference path Pr is calculated. a is the risk value R RMAs long as the reference path Pr is greater than the reference path Pc, the reference path Pr is set as the selected path Pc.
[0069] 6, when the vehicle passes by the utility pole Z2 on the right side on the same road, the allowable risk U corresponding to the current vehicle speed V (=30 km / h) on the acquired reference path Pr is a is the risk value R RM If the risk is less than a is the risk value R RM Of the above possible travel routes, the position closest to the reference path Pr is set as the selected path Pc.
[0070] 7, when a vehicle passes by a pedestrian Y1 on the right side on the same road, the risk value R RM is set high, even if the utility pole Z2 shown in FIG. 6 and the pedestrian Y1 shown in FIG. 7 are at the same distance from the vehicle, the allowable risk U corresponding to the current vehicle speed V (= 30 km / h) is a is the risk value R RM In such a case, the vehicle speed V of the host vehicle is reduced (V=20 km / h) to reduce the allowable risk U a increases, and the acceptable risk U a is the risk value R RM The selected path Pc is set by creating the above-described drivable route.
[0071] (Method of the Present Embodiment Reflecting the Tolerable Risk for Each Driver) Up to this point, the basic method of setting the driving conditions by the driving condition setting unit 71 has been described. Here, as shown in FIG. 8 , in a scene where the vehicle 1 is passing by a parked vehicle 100 parked on the road in the first lane (own lane) in which the vehicle 1 is traveling, there is considered to be a driver who senses danger (collision risk) in contact with the parked vehicle 100 and selects a trajectory 81 that crosses the boundary line (center line) 101 and significantly protrudes into the second lane (oncoming lane) adjacent to the first lane. It is also considered that there is a driver who feels reluctant to deviate from the rules (rule deviation risk) by crossing the boundary line 101 and protruding into the oncoming lane, and selects a trajectory 83 that minimizes the amount of protrusion. In the basic method described above, the tolerable risk Ua Since there is only one calculation formula for the driver, it is not possible to reflect the concept of tolerance for the risk of deviation from the rules.
[0072] In contrast to this, in this embodiment, risk targets are divided into collision risk, rule deviation risk, and road departure risk, and for each risk, an acceptable collision risk, rule deviation risk, and road departure risk are set that reflect the driver's acceptable risk. Note that in this embodiment, rule deviation risk and road departure risk are distinguished as being different departure risks between crossing the boundary line 101 into an oncoming lane and crossing the road edges 103, 105 onto a sidewalk or shoulder. However, these may also be regarded as a common deviation risk, and a common deviation allowable risk may be set.
[0073] 9 to 12 are diagrams for explaining the risk potential set in the scene shown in Fig. 8. Fig. 9 shows the position of the parked vehicle 90 in the scene shown in Fig. 8 as seen from the direction of travel of the vehicle 1. Figs. 10 to 12 each show the collision risk R set for the parked vehicle 90. col , the rule deviation risk R set for the boundary line 101 rul , and the road departure risk R dev 1 is an explanatory diagram of the vehicle 1 viewed in the direction of travel.
[0074] The risk setting unit 67 calculates a collision risk R col , rule deviation risk R rul and road departure risk R dev The collision risk R col , the risk value increases as the vehicle approaches the parked vehicle 90, and the risk value is at its maximum within the range in which the parked vehicle 90 exists. rul In the range close to the boundary line 101, the risk value increases as the vehicle approaches the boundary line 101, and the risk value gradually increases as the vehicle moves beyond the boundary line 101 and away from the boundary line 101.dev In the example, the risk value increases as the vehicle approaches the road edges 103 and 105, and reaches a maximum at the road edges 103 and 105.
[0075] These collision risks R col , rule deviation risk R rul and road departure risk R dev The value of may be a value set in advance depending on the risk object. For example, the value of the collision risk R col and road departure risk R dev may be set using the above formula (1). rul The risk value of the range of the current lane may be set using the above formula (1). rul The risk value for the oncoming lane range may be set at any gradient.
[0076] In particular, in this embodiment, the rule deviation risk R rul is set according to the strength of the traffic rule. The "strength of the traffic rule" indicates the strength of the constraint of the traffic rule that must be observed or the level of danger that occurs when a traffic rule is violated. For example, in the example of the boundary line 101 of a lane, the tolerance for crossing the boundary line 101 decreases in the order of dashed line, white line, and orange line, and therefore the strength of the traffic rule increases. In addition, the strength of the traffic rule for a restricted driving zone such as a zebra zone (guide zone) can also be defined depending on the type. For example, a zebra zone does not prohibit the entry of a vehicle 1, but can be considered an area where entry is undesirable. However, some zebra zones, for example, are areas surrounded by an orange border indicating no entry. For such restricted driving zones, the tolerance for a vehicle 1 entering the restricted driving zone varies depending on the type, and the strength of the traffic rule can also be defined.
[0077] In this embodiment, the rule deviation risk R rul The strength of the traffic rule is set in advance according to the type of risk object to be set, and the maximum value C of the risk value in the above formula (1) is set according to the type of risk object and the strength of the traffic rule. i or risk target range r iand a gradient of the risk value of the range of the destination where the vehicle 1 crosses the boundary line 101 or the boundary of the travel restriction zone. For example, the higher the strength of the traffic rules, the higher the maximum risk value C i Or risk range r i is set large, and furthermore, the gradient of the risk value of the range of the destination where the vehicle 1 crosses the boundary line 101 or the boundary of the travel restriction zone is set large. This makes it possible to set a target trajectory that reflects the strength of the traffic rules.
[0078] 13 shows an example in which the strength of traffic rules is set for each type of boundary line 101 and traffic regulation zone. Note that the boundary line 101 includes not only the boundary line (center line) between the current lane (first lane) and the oncoming lane (second lane), but also the boundary line between the current lane (first lane) and the adjacent lane (second lane) on roads with two or more lanes in each direction.
[0079] In the setting example shown in Fig. 13, the intensity of the traffic rules is set to "low" when the boundary line 101 is a dashed line. The intensity of the traffic rules is set to "medium" when the boundary line 101 is a white line and when the traffic regulation zone is a zebra stripe zone. The intensity of the traffic rules is set to "high" when the boundary line 101 is an orange line and when the traffic regulation zone is a no-entry zone.
[0080] FIG. 14 shows the rule deviation risk R rul 14 is an explanatory diagram showing an example in which the rule deviation risk R rul_wh and the rule deviation risk R when the boundary line 101 is an orange line rul_or The orange line has stronger restrictions on going outside the line than the white line, so the rule deviation risk R rul_or The risk value (maximum value) on the boundary line 101 is the rule deviation risk R rul_wh The risk value (maximum value) on the boundary line 101 is larger than the risk value on the boundary line 101, and the risk range is also set to be larger.rul_or The gradient of the risk value on the oncoming lane side is the rule deviation risk R rul_wh The gradient of the risk value on the opposite side of the lane is set to be larger than that of the risk value on the opposite side of the lane.
[0081] 15 and 16 show the rule deviation risk R rul_zz 15 is an explanatory diagram showing an example of a zebra zone, which is a guidance zone just before the point where the lane for vehicle 1 widens to two lanes, and FIG. rul_zz The illustrated rule deviation risk R rul_zz In the example, the risk value increases as the vehicle approaches the boundary between the vehicle's lane and the zebra zone, and remains constant within the zebra zone. Furthermore, in the area beyond the zebra zone toward the oncoming lane, the risk value increases even more as the vehicle moves away from the zebra zone. This indicates that the risk value increases as the vehicle moves further into the oncoming lane.
[0082] Next, as shown in FIGS. 17 to 19, the tolerable risk setting unit 69 determines the collision risk R col , rule deviation risk R rul and road departure risk R dev Collision tolerance risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev Set the allowable collision risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev are respectively represented by the following formulas (3) to (5).
[0083]
[0084] U col : Collision tolerance risk w col : Gain V: Vehicle speed
[0085]
[0086] U rul : Acceptable risk of deviation from rules w rul : Gain V: Vehicle speed
[0087]
[0088] U dev : Collision tolerance risk w dev : Gain V: Vehicle speed
[0089] Collision tolerance risk U col and road departure allowable risk U dev is inversely proportional to the square of the speed V of the vehicle 1. Therefore, when setting a trajectory that approaches an obstacle or the edge of the road, the acceptable collision risk U can be reduced by reducing the vehicle speed V. col and road departure allowable risk U dev Therefore, careful driving when approaching an obstacle or driving on a narrow road is reproduced. On the other hand, the allowable risk of deviation from the rules, U rul is proportional to the speed of the vehicle 1. Therefore, when a trajectory that strays into the oncoming lane must be set, the greater the amount of straying into the oncoming lane, the greater the vehicle speed V is, thereby reducing the allowable risk of deviation from the rule. rul Therefore, the driving behavior in which the state where the boundary line 101 is exceeded is quickly resolved is reproduced.
[0090] 20 and 21 are explanatory diagrams showing differences in the permissible driving range depending on the permissible risk. col and rule deviation tolerance risk U rul FIG. 21 shows the tolerable collision risk U that is set when the vehicle 1 is traveling at a relatively high speed. col and rule deviation tolerance risk U rul For ease of understanding, in the examples shown in FIGS. 20 and 21, the road departure risk R dev and road departure allowable risk U dev is omitted from the illustration.
[0091] As shown in FIG. 20, when the vehicle 1 is traveling at a low speed, the acceptable collision risk U col is the collision risk R col and the allowable risk of deviation from the rules U rul is the risk of rule violation R rulThe permissible driving range Px exceeding the limit is limited to the range of the own lane close to the parked vehicle 90. On the other hand, as shown in FIG. 21, when the vehicle 1 is traveling at high speed, the permissible collision risk U col is the collision risk R col and the allowable risk of deviation from the rules U rul is the risk of rule violation R rul The allowable driving range Py exceeding this range is a range that includes the oncoming lane away from the parked vehicle 90.
[0092] The acceptable collision risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev is learned by the learning processing unit 65 based on data obtained when the driver manually drives the vehicle in each scene (driving scene). For example, the learning processing unit 65 calculates the collision risk R based on information about the surrounding environment of the vehicle 1 and data about the vehicle's operation state and behavior collected in a scene in which the vehicle 1 passes by a parked vehicle 90 shown in FIG. col , rule deviation risk R rul and road departure risk R dev and Collision Tolerance Risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev The least squares fitting process is performed using the equations (3) to (5) and the gain W col , W rul , W dev Learn.
[0093] Specifically, the learning processing unit 65 identifies a driving scene in which the driver operated the steering wheel, accelerator pedal, or brake pedal to maintain a certain distance from an obstacle, based on data on the surrounding environment of the vehicle 1 detected during manual driving and data on the operation state and behavior of the vehicle 1. The learning processing unit 65 receives as input data the distance between the vehicle 1 and an obstacle, etc., the vehicle speed, and the amount of protrusion (protrusion amount) from a boundary line (center line) in the identified driving scene. In this case, the distance between the vehicle 1 and the obstacle, etc., is the distance between the vehicle 1 and the obstacle, etc., in a direction perpendicular to the traveling direction of the vehicle 1 when the vehicle 1 passes by the obstacle. In this case, the distance between the vehicle 1 and the obstacle, etc., also includes the distance from the vehicle 1 to the edge of the road.
[0094] The learning processing unit 65 calculates the tolerable collision risk U based on the specified input data. col , Rule deviation tolerance risk U rul and road departure allowable risk U dev The gain W of equations (3) to (5) col , W rul , W dev This allows a driver who feels a strong resistance to drifting into the oncoming lane and a driver who feels a great risk of hitting a parked vehicle 90 to set an allowable driving range that is acceptable for each. At this time, the learning processing unit 65 distinguishes the type of boundary line 101 or the type of driving restriction zone, and calculates the rule deviation risk R set for each risk object. rul Tolerance risk of deviation from rules rul Therefore, it is possible to set an acceptable driving range that reflects the driver's acceptable risk, which may change depending on the strength of traffic rules.
[0095] 22 and 23 show the situation shown in FIG. 8 in which the parked vehicle 90 is at a collision risk R col is set, and the rule deviation risk R rul In the example in which the above values are set, the permissible driving ranges Pa and Pb are respectively set for driver A who feels a strong resistance to straying into the oncoming lane and driver B who feels a great risk of hitting a parked vehicle 90.
[0096] In the case of driver A, who feels a strong resistance to going into the oncoming lane, during manual driving, the driver A decelerates and passes beside the parked vehicle 90 so as to minimize the amount of going into the oncoming lane. In this case, the acceptable collision risk U col is set to a relatively large value corresponding to low speed driving, and the allowable risk of deviation from the rules U rul is set to a relatively small value corresponding to low-speed driving. Therefore, the speed of the vehicle 1 when passing beside the parked vehicle 90 by automatic driving is set to a relatively low speed, and the driving trajectory is set so as to pass through the permissible driving range Pa within the own lane.
[0097] On the other hand, in the case of driver B who feels a high risk of hitting the parked vehicle 90, when manually driving, the driver passes beside the parked vehicle 90 so as to extend significantly into the oncoming lane. In this case, the collision tolerance risk U col is set to a relatively small value corresponding to high-speed driving, and the allowable risk of deviation from the rules U rul is set to a relatively large value corresponding to high-speed driving. Therefore, the speed of vehicle 1 when passing by parked vehicle 90 by automatic driving is set to a relatively high speed, and the driving trajectory is set to pass through permissible driving range Pb in the oncoming lane. However, because the vehicle speed is set to a high speed, the time that vehicle 1 travels in the oncoming lane is shortened, and vehicle 1 is quickly returned to its own lane.
[0098] In addition, for drivers who feel a strong resistance to both approaching obstacles and violating the rules, the acceptable collision risk U col and rule deviation tolerance risk U rul In this case, as shown in FIG. 24, the acceptable collision risk U col is the collision risk R col and the allowable risk of deviation from the rules U rul is the risk of rule violation R rul . . . may not be formed.
[0099] In such a case, as shown in Fig. 25, the driving condition setting unit 71 increases the vehicle speed to set a driving trajectory on the side of the oncoming lane in order to create an allowable driving range, and by quickly passing through the oncoming lane, the time for the rule deviation state to be shortened. dev is set, and the road departure risk R dev Tolerable risk of road departure U dev is set. The allowable risk of road departure U dev The faster the vehicle speed, the smaller the value becomes, and the allowable driving range of the vehicle 1 is moved away from the road edge. Therefore, the allowable road departure risk U dev This makes it possible to limit the amount of vehicle 1 protruding into the oncoming lane.
[0100] In this way, the driving condition setting unit 71 sets driving conditions that reflect not only the physical risk level of an obstacle that the driver perceives as risky, but also the acceptable collision risk, acceptable rule deviation risk, and acceptable road departure risk for each driver. Therefore, appropriate automated driving control can be executed in various driving situations according to the risk that each driver perceives for detected obstacles. This realizes safe automated driving for each driver and increases the reliability of the automated driving control.
[0101] (1-3-2. Processing Operation) Next, an example of the processing operation of the driving assistance device 50 will be described, divided into a learning process and a driving assistance process, using the case of setting driving conditions when a vehicle passes beside a parked vehicle as described above as an example. Note that in the following description, detailed explanations of parts that overlap with the contents explained so far will be omitted.
[0102] (1-3-2-1. Learning Process) FIG. 26 is a flowchart showing the learning process.
[0103] First, the learning processing unit 65 detects the driver's face based on image data transmitted from the in-vehicle camera 21, and generates identification information for distinguishing the driver by analyzing facial features or the like, and sets the information in the learning data storage unit 55 (step S11). As a result, data to be learned from now on is linked to the specific identification information, and can be identified as learning data for that driver. The identification information for distinguishing the driver may be input by the driver.
[0104] Next, the learning processing unit 65 determines whether the driver is manually driving the vehicle 1 (step S13). For example, the learning processing unit 65 acquires a signal or message indicating the status of the driving mode from the vehicle control device 41 and determines whether the vehicle 1 is manually driven. If the learning processing unit 65 does not determine that the driver is manually driving the vehicle 1 (S13 / No), it repeats the determination of step S13 until it determines that the vehicle 1 is manually driven.
[0105] On the other hand, if the learning processing unit 65 determines that the driver is manually driving the vehicle 1 (S13 / Yes), it acquires information about the surrounding environment of the vehicle 1 detected by the surrounding environment detection unit 63 and data on the operation state and behavior of the vehicle 1 detected by the vehicle state sensor 25 (step S15).
[0106] Next, the learning processing unit 65 determines whether the vehicle 1 has corrected its traveling trajectory to ensure a sufficient distance from the obstacle, based on the acquired information on the surrounding environment of the vehicle 1 and the data on the operation state / behavior (step S17). For example, the learning processing unit 65 determines that the vehicle 1 has corrected its traveling trajectory to ensure a sufficient distance from the obstacle when, before or after passing by an obstacle in the traveling lane (own lane) of the vehicle 1, the vehicle 1 changes its traveling position in the traveling lane or performs a steering operation to go around the obstacle. However, the method for determining whether the vehicle 1 has corrected its traveling trajectory to ensure a sufficient distance from the obstacle is not particularly limited.
[0107] If the learning processing unit 65 does not determine that the vehicle 1 has corrected its traveling trajectory to maintain a sufficient distance from the obstacle (S17 / No), the process returns to step S13 and repeats the processing of each of the above steps. On the other hand, if the learning processing unit 65 determines that the vehicle 1 has corrected its traveling trajectory to maintain a sufficient distance from the obstacle (S17 / Yes), the learning processing unit 65 records information about the surrounding environment of the vehicle 1 and data on the operation state / behavior when the vehicle 1 corrected its traveling trajectory (step S19). Specifically, the learning processing unit 65 records data on the distance between the vehicle 1 and the obstacle, the type of obstacle, the speed of the vehicle 1, the type of boundary line or travel restriction zone, the amount by which the vehicle 1 has protruded from the boundary line (protrusion amount) from the vehicle's own lane, the amount by which the vehicle 1 has entered the travel restriction zone (protrusion amount) from the vehicle's own lane, and the distance from the vehicle 1 to the road edge when the vehicle 1 passes by the obstacle. The recorded data may include information other than the above.
[0108] Next, the learning processing unit 65 determines whether a certain amount of data has been accumulated (step S21). For example, the learning processing unit 65 determines whether the number of times that the vehicle 1 has determined that it has corrected its traveling trajectory to ensure a sufficient distance from an obstacle and recorded data has reached a predetermined number. However, the learning processing unit 65 may determine that a certain amount of data has been accumulated after a certain period of time has elapsed, or may determine that a certain amount of data has been accumulated for each driving cycle from the start to the end of driving the vehicle 1.
[0109] If the learning processing unit 65 does not determine that a certain amount of data has been accumulated (S21 / No), the process returns to step S13 and repeats the processing of each step described above. On the other hand, if the learning processing unit 65 determines that a certain amount of data has been accumulated (S21 / Yes), the learning processing unit 65 determines that the tolerable collision risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev Specifically, the collision risk R is calculated based on the accumulated information on the surrounding environment of the vehicle 1 and the data on the operation state / behavior of the vehicle 1. col , rule deviation risk R rul and road departure risk R dev and Collision Tolerance Risk U col, Rule deviation tolerance risk U rul and road departure allowable risk U dev The least squares fitting process is performed using the equations (3) to (5) and the gain W col , W rul , W dev Learn.
[0110] At this time, the learning processing unit 65 calculates the allowable risk of deviation from the rule U for each type of boundary line 101 that existed around the vehicle 1 when the vehicle 1 passed by the obstacle and for each type of traffic regulation zone. rul The gain W of the formula (4) rul Learn.
[0111] Next, the learning processing unit 65 calculates the learned collision tolerance risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev The learning processing unit 65 stores the equations (3) to (5) in association with the driver's identification information in the learning data storage unit 55 (step S25). rul The equation (4) is stored in the learning data storage unit 55.
[0112] In this way, the learning processing unit 65 learns the tolerable risk that each driver perceives for each risk object based on data collected during the period when each driver manually drives the vehicle, and stores the learning result data in the learning data memory unit 55.
[0113] (1-3-2-2. Driving Assistance Processing) Figures 27 to 30 are flowcharts for explaining the operation of the driving assistance processing. Figure 27 shows the main routine of the driving assistance processing.
[0114] First, when the driving system of the vehicle 1 is activated (step S31), the driving condition setting unit 71 executes a process of identifying the driver of the vehicle 1 (step S33). Specifically, the driving condition setting unit 71 detects the driver's face based on image data transmitted from the in-vehicle camera 21, and identifies the driver and the driver's identification information by analyzing facial features or the like. This makes it possible to identify the learning data to be referenced. Note that the information identifying the driver currently performing driving operations may be information input by the driver.
[0115] Next, the driving condition setting unit 71 determines whether the vehicle 1 is in autonomous driving or not (step S35). For example, the driving condition setting unit 71 acquires a signal or message indicating the status of the driving mode from the vehicle control device 41 and determines whether the vehicle 1 is in autonomous driving or not. If the driving condition setting unit 71 does not determine that the vehicle 1 is in autonomous driving or not (S35 / No), it repeats the determination of step S35.
[0116] On the other hand, if the driving condition setting unit 71 determines that the vehicle 1 is in autonomous driving (S35 / Yes), it acquires information about the road in the traveling direction of the vehicle 1 (step S37). Specifically, the driving condition setting unit 71 acquires information about the position and type of boundary line between the first lane in which the vehicle 1 is traveling and the adjacent second lane, the position and type of driving restriction zone, and the position and type of road edge, based on the information about the surrounding environment detected by the surrounding environment detection unit 63. The type of road edge indicates a boundary line, curb, guardrail, or the like that defines the road edge.
[0117] Next, the risk setting unit 67 and the tolerable risk setting unit 69 set a risk potential and a tolerable risk for the risk object included in the acquired road information (step S39). Figure 28 shows a flowchart of the process for setting a risk potential and a tolerable risk for a risk object on a road.
[0118] First, the risk setting unit 67 calculates a rule deviation risk R that is set in advance according to the type of risk object for each risk object on the road. rul and road departure risk R devSpecifically, the risk setting unit 67 sets a rule deviation risk R that is set in advance according to the type of each of the detected boundary lines, restricted driving zones, and road edges. rul and road departure risk R dev In particular, the risk setting unit 67 determines the type of boundary line or restricted driving zone, and sets a rule deviation risk R that is set in advance according to the strength of the traffic rule corresponding to the type of boundary line or restricted driving zone. rul Set.
[0119] Next, the allowable risk setting unit 69 sets an allowable risk for each of the set risk potentials (step S53). Specifically, the allowable risk setting unit 69 calculates the rule deviation risk R from among the equations (3) to (5) that indicate the allowable risk and are stored in the learning data storage unit 55 in association with the driver's identification information. rul and road departure risk R dev For each risk object for which a rule deviation allowable risk U is set, rul and road departure allowable risk U dev Furthermore, the allowable risk setting unit 69 selects the formulas (4) to (5) that indicate the selected rule deviation allowable risk U rul and road departure allowable risk U dev The formulas (4) and (5) are used to calculate the corresponding rule deviation risk R rul and road departure risk R dev Set the values for each.
[0120] 27 , the driving condition setting unit 71 then acquires information about obstacles in the traveling direction of the vehicle 1 (step S41). Specifically, the driving condition setting unit 71 acquires information about oncoming vehicles, parked vehicles, pedestrians, bicycles, buildings, telephone poles, traffic lights, and other three-dimensional objects based on the information about the surrounding environment detected by the surrounding environment detection unit 63. The obstacle information includes at least data about the type of obstacle, the size of each obstacle, the distance from the obstacle, and the relative speed from the obstacle.
[0121] Next, the risk setting unit 67 and the tolerable risk setting unit 69 set a risk potential and a tolerable risk for the detected obstacle (step S43). Figure 29 shows a flowchart of the process for setting a risk potential and a tolerable risk for an obstacle.
[0122] First, the risk setting unit 67 calculates a collision risk R for each obstacle, which is set in advance according to the type of the obstacle. col Next, the tolerable risk setting unit 69 sets each of the set collision risks R col Collision tolerance risk U col Specifically, the tolerable risk setting unit 69 selects the collision risk R from among the equations (3) to (5) that indicate the tolerable risk and are stored in the learning data storage unit 55 in association with the driver's identification information (step S63). col The collision tolerance risk U col Furthermore, the allowable risk setting unit 69 selects the equation (3) that indicates the selected allowable collision risk U col The equation (3) showing the corresponding collision risk R col Set the values for each.
[0123] 27, next, the driving condition setting unit 71 sets a target trajectory and a target speed of the vehicle 1 (step S45). Fig. 30 shows a flowchart of the process for setting the target trajectory and the target speed.
[0124] First, the driving condition setting unit 71 defines cross sections perpendicular to the route at each point set at a regular interval on the reference path, which is the planned driving route of the vehicle 1 (step S71). Specifically, the driving condition setting unit 71 acquires information on the current position of the vehicle 1 output from the GNSS sensor 27 and also refers to the map data storage unit 29 to acquire data on the reference path, which is the driving route that serves as a reference when driving on the road currently being driven. Furthermore, the driving condition setting unit 71 defines cross sections perpendicular to the route at the position of each point set on the reference path. Each cross section is assigned a collision risk R col , rule deviation risk R rul and road departure risk R dev(See FIG. 22, etc.) The intervals between points set on the reference path may be a fixed distance interval, or may be a distance interval corresponding to a fixed traveling time at the current vehicle speed.
[0125] Next, the driving condition setting unit 71 determines whether or not there is a range in each cross section where the vehicle 1 can travel at the current speed (step S73). col , rule deviation risk R rul and road departure risk R dev Using the equations (3) to (5) set for the current vehicle speed V, the acceptable collision risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev As mentioned above, the acceptable collision risk U col and road departure allowable risk U dev The larger the vehicle speed V, the smaller the value becomes, and the allowable risk of deviation from the rules U rul The higher the vehicle speed V, the larger the value of
[0126] The driving condition setting unit 71 calculates the collision tolerance risk U col is the collision risk R col The allowable risk of deviation from the rules is U rul is the risk of rule violation R rul and the acceptable risk of road departure U dev is the risk of road departure R dev It is determined whether there is a region (permissible driving range) exceeding the above.
[0127] When the driving condition setting unit 71 determines that there is a range in each cross section where the vehicle 1 can travel at the current speed (S73 / Yes), it sets the position of the allowable driving range in each cross section that is closest to a point on the reference path as the target trajectory in that cross section (step S79). In this case, the driving condition setting unit 71 sets the current vehicle speed V as the target speed. This allows the vehicle 1 to travel on a driving trajectory that poses a low risk to the driver while maintaining the current vehicle speed V.
[0128] On the other hand, when the driving condition setting unit 71 does not determine that a range in which the vehicle 1 can travel at the current speed exists at each cross section (S73 / No), that is, when a range in which the vehicle 1 can travel at the current speed does not exist at any cross section, the driving condition setting unit 71 increases or decreases the speed at a fixed interval from the current vehicle speed V (step S75). Here, the driving condition setting unit 71 does not increase or decrease the actual vehicle speed V, but increases or decreases the assumed vehicle speed. The interval by which the vehicle speed is increased or decreased is set to an arbitrary value (e.g., 5 km / h) in advance. The interval by which the vehicle speed is increased or decreased may be variable depending on the current vehicle speed V, or may be set to different values when the vehicle speed is increased and when it is decreased. Whether the vehicle speed is increased or decreased may be determined, for example, based on whether the driving range is increased or decreased.
[0129] Next, the driving condition setting unit 71 determines whether or not there is a range in which the vehicle can travel at the increased or decreased vehicle speed (step S77). col , rule deviation risk R rul and road departure risk R dev Using the equations (3) to (5) set for the vehicle speed V after the increase or decrease, the collision tolerance risk U col , Rule deviation tolerance risk U rul and road departure allowable risk U dev Calculate.
[0130] The driving condition setting unit 71 calculates the collision tolerance risk U col is the collision risk R col The allowable risk of deviation from the rules is U rul is the risk of rule violation R rul and the acceptable risk of road departure U dev is the risk of road departure R dev It is determined whether there is a region (permissible driving range) exceeding the above.
[0131] If the driving condition setting unit 71 does not determine that a range in which the vehicle can be driven at the increased or decreased vehicle speed exists (S77 / No), the process returns to step S75, and the driving condition setting unit 71 repeats increasing or decreasing the vehicle speed until it can determine that a range in which the vehicle can be driven at the increased or decreased vehicle speed exists. If the driving condition setting unit 71 determines that a range in which the vehicle can be driven at the increased or decreased vehicle speed exists (S77 / Yes), the driving condition setting unit 71 sets the position of the permissible driving range at each cross section that is closest to a point on the reference path as the target trajectory at that cross section (step S79). In this case, the driving condition setting unit 71 sets the current vehicle speed V after the increase or decrease as the target speed. This allows the vehicle to be driven at a vehicle speed V and on a driving trajectory that poses a low risk to the driver.
[0132] 27 , the driving condition setting unit 71 outputs information about the set target trajectory and target speed to the vehicle control device 41 (step S47). Having acquired the information about the target trajectory and target speed, the vehicle control device 41 sets control target values for the acceleration / deceleration and steering angular velocity of the vehicle 1 based on the information about the target trajectory and target speed, and controls the driving of the driving force source 9, the electric steering device 15, and the brake control unit 16. As a result, the vehicle 1 travels by automatic driving on a traveling trajectory and at a vehicle speed that reflects the driver's tolerable risk.
[0133] Next, the driving condition setting unit 71 determines whether the driving system of the vehicle 1 has stopped (step S49). If the driving condition setting unit 71 does not determine that the driving system of the vehicle 1 has stopped (S49 / No), the driving condition setting unit 71 returns to step S35 and repeats the processing of each of the above-mentioned steps. On the other hand, if the driving condition setting unit 71 determines that the driving system of the vehicle 1 has stopped (S49 / Yes), the driving assistance processing routine is terminated.
[0134] <1-4. Effects of the driving assistance device according to this embodiment> As described above, the driving assistance device 50 according to this embodiment sets a rule deviation risk indicating the risk of deviating from traffic rules related to the driving of the vehicle 1, in addition to the collision risk perceived by the driver with surrounding obstacles, and sets an acceptable collision risk for the collision risk and an acceptable rule deviation risk for the rule deviation risk, thereby setting driving conditions for automated driving of the vehicle 1. The rule deviation risk and acceptable rule deviation risk that are set are set respectively for risk targets with different traffic rule strengths.
[0135] Therefore, when changing the driving trajectory to ensure a sufficient distance between the vehicle 1 and an obstacle, the driving conditions can be set according to the strength of the traffic rule. For example, the stronger the traffic rule, the greater the risk of violating the rule, and the priority is given to actions that reduce the risk of violating the rule. Therefore, a driving assistance system that can give drivers a sense of security and trust can be realized, and each driver will continue to use the driving assistance system.
[0136] Furthermore, the allowable risk of deviation from the rules is set to increase in proportion to the vehicle speed. Therefore, when a target trajectory that deviates greatly from the rules is set to ensure a large distance from an obstacle, the target vehicle speed is set to a large value, thereby enabling driving that can shorten the duration of the rule deviation state. Furthermore, the allowable risk of collision is set to decrease in proportion to the vehicle speed. Therefore, when a target trajectory that is close to an obstacle is set, the target vehicle speed is set to a small value, allowing safe and secure driving.
[0137] The set acceptable collision risk is set based on a calculation formula learned from data collected during manual driving by the driver, and reflects the risk perceived by each driver in relation to the distance to an obstacle and the passing speed. The set acceptable rule deviation risk is set based on a calculation formula learned according to the strength of traffic rules, and reflects the risk perceived by each driver in relation to deviating from traffic rules.
[0138] Therefore, when changing the driving trajectory to ensure a sufficient distance between the vehicle 1 and an obstacle, it is possible to set driving conditions that reflect the risk perceived by the driver depending on the obstacle and the severity of the traffic rule. For example, the higher the risk perceived by the driver in violating a traffic rule, the smaller the tolerable risk of rule violation, and priority is given to actions that reduce the risk of rule violation. Therefore, it is possible to realize a driving assistance system that can provide a sense of security and trust in accordance with the characteristics of each driver. This allows each driver to continuously use the driving assistance system.
[0139] Furthermore, in this embodiment, a road departure risk is set separately from the collision risk and the rule deviation risk, and the allowable road departure risk, which is set to decrease in proportion to the vehicle speed, is used to set the driving conditions of the vehicle 1. Therefore, it is possible to set driving conditions that reflect the risk that the driver feels when the vehicle 1 goes off the road edge and enters a sidewalk or shoulder strip.
[0140] Furthermore, according to the driving assistance device 50 of this embodiment, the learning processing unit 65 uses data collected during manual driving of the vehicle to learn equations that indicate the acceptable collision risk, the acceptable rule deviation risk, and the acceptable road departure risk, and stores the learned data in the learned data storage unit 55. Therefore, learned data is automatically accumulated while the driver is manually driving the vehicle 1. Therefore, learned data of characteristics that appear during manual driving by the driver and affect risk perception is accurately accumulated, and the accuracy of the set control conditions can be improved.
[0141] <<2. Second embodiment>> Next, a driving assistance device 50 according to a second embodiment of the present disclosure will be described. In the first embodiment, the tolerance for crossing a lane boundary line or the tolerance for entering a restricted driving zone is defined as the strength of a traffic rule, and a rule deviation risk and a rule deviation tolerance risk are set according to the strength of the traffic rule, and driving conditions for the vehicle 1 are set. In the second embodiment, an example that can be applied to a situation in which the strength of a traffic rule transitions spatially on the driving route of the vehicle 1 will be described.
[0142] 31 is an explanatory diagram showing a scene to which an example of this embodiment can be applied. Fig. 31 shows a scene in which a vehicle 1a to be assisted, traveling in the first lane (driving lane) of a two-lane road, overtakes another vehicle 91a, also traveling in the first lane at a low speed. The positions of vehicle 1a and the other vehicle 91a, vehicle 1b and the other vehicle 91b, and vehicle 1c and the other vehicle 91c, respectively, show the positions of vehicle 1a and the other vehicle 91a at the same time.
[0143] There is an intersection ahead of vehicle 1a, with lane 1 allowing vehicles to continue straight through the intersection and lane 2 (passing lane) adjacent to lane 1 allowing vehicles to turn right at the intersection. The boundary between lane 1 and lane 2 changes from a dashed white line 101a to an orange line 101b just before the intersection. In other words, the boundary between lane 1 and lane 2 changes along the direction of travel of vehicle 1a, and the strength of the traffic rules changes spatially.
[0144] In such a situation, the driving assistance device 50 of the vehicle 1 sets the tolerable risk for the second lane so that the overtaking will be completed before the boundary line changes from the broken white line 101a to the orange line 101b. Specifically, the driving assistance device 50 sets the overtaking completion point, for example, at the point where the boundary line changes from the broken white line to the orange line, and calculates the target overtaking route. This allows the distance that the vehicle 1a will travel in the second lane from when it enters the second lane until it leaves the second lane to be calculated.
[0145] Furthermore, the driving assistance device 50 calculates a target vehicle speed for the vehicle 1a to travel along the overtaking route without contacting the other vehicle 91a, based on the travel distance in the second lane and the speed of the other vehicle 91a. That is, the target vehicle speed while traveling in the second lane is calculated so that the vehicle 1a can return to the first lane without contacting the other vehicle 91a. The target vehicle speed calculated here is substituted into equation (4) indicating the allowable risk of deviation from the rule, thereby determining the allowable risk of deviation from the rule in the second lane.
[0146] This embodiment can also be applied to a situation where there is a blind spot on either the left or right side due to a curve in the road ahead in the direction of travel, and another vehicle 91a is traveling at a low speed ahead of vehicle 1a, by using the lane on the opposite side of the direction in which the blind spot exists as the first lane in the above example.
[0147] FIG. 32 shows a flowchart of the process for setting the risk potential and the tolerable risk according to this embodiment.
[0148] First, the risk setting unit 67 calculates a rule deviation risk R that is set in advance according to the type of risk object for each risk object on the road, in the same procedure as in step S51 of the flowchart shown in FIG. rul and road departure risk R dev is set (step S51).
[0149] Next, the risk setting unit 67 determines whether or not the situation is one in which the vehicle 1a is overtaking a moving obstacle (step S54). If the risk setting unit 67 determines that the situation is not one in which the vehicle 1a is overtaking a moving obstacle (S54 / No), the process proceeds to step S53. In step S53, the allowable risk setting unit 69 sets an allowable risk for each of the set risk potentials in the same manner as in step S51 of the flowchart shown in FIG. 28 (step S53). In other words, this case is one in which the vehicle 1a is overtaking a stationary obstacle, and the risk potential and allowable risk are set in accordance with the flowchart shown in FIG. 28.
[0150] On the other hand, if the risk setting unit 67 determines that the situation is one in which the vehicle 1a is about to overtake a moving obstacle (another vehicle 91a) (S54 / Yes), the allowable risk setting unit 69 determines whether or not there is a no-overtaking point or a point with poor visibility at a point where the vehicle 1a will return to the current driving lane (first driving lane) after overtaking (step S55). For example, the allowable risk setting unit 69 determines whether or not there is a no-overtaking point or a point with poor visibility at a point where the vehicle 1a will return to the current driving lane after overtaking, based on the information about the surrounding environment output from the surrounding environment sensor 23, or based on the position information and map data of the vehicle 1a acquired by the GNSS sensor 27.
[0151] If the allowable risk setting unit 69 does not determine that an overtaking no-passing point or a point with poor visibility exists at the point where the vehicle 1a will return to the current driving lane after overtaking (S55 / No), the allowable risk setting unit 69 proceeds to step S53 and sets an allowable risk for each of the set risk potentials, as in the case of overtaking a stationary obstacle.On the other hand, if the allowable risk setting unit 69 determines that an overtaking no-passing point or a point with poor visibility exists at the point where the vehicle 1a will return to the current driving lane after overtaking (S55 / Yes), the allowable risk setting unit 69 calculates the speed required to complete the overtaking by the point based on the distance to the point and the traveling speed of the object to be overtaken (step S56).
[0152] Specifically, the allowable risk setting unit 69 sets a point where the vehicle 1a can complete the overtaking without coming into contact with an obstacle before any overtaking no-passing point or point with poor visibility, and calculates a target overtaking route. The allowable risk setting unit 69 also calculates the distance that the vehicle 1 will travel in the second lane adjacent to the driving lane during the overtaking route, from when the vehicle 1 enters the second lane until when the vehicle 1 leaves the second lane. Furthermore, the allowable risk setting unit 69 calculates the required speed of the vehicle 1a for traveling along the overtaking route without coming into contact with the other vehicle 91a, based on the distance that the vehicle 1a travels in the second lane and the speed of the other vehicle 91a.
[0153] Next, the allowable risk setting unit 69 adjusts the gain W in the equation (4) that indicates the allowable risk of deviation from the rule so that the target speed of the vehicle 1a becomes the required speed.rul is adjusted (step S57).
[0154] Next, the allowable risk setting unit 69 sets an allowable risk for each of the set risk potentials (step S53). At this time, the allowable risk R rul Tolerance risk of deviation from rules rul is the gain W calculated in step S57 rul This makes it possible to set driving conditions that allow the vehicle to complete overtaking before the no-overtaking point or the point with poor visibility, even when the vehicle is following a route that involves entering an adjacent lane to overtake a moving obstacle.
[0155] As described above, according to the example of this embodiment, in a situation where the strength of a traffic rule spatially transitions on the travel route of the vehicle 1, it is possible to anticipate the transition in the strength of the traffic rule and set driving conditions that enable the vehicle to complete overtaking before the transition in the traffic rule. Therefore, it is possible to realize a driving assistance system that can give drivers a sense of security and trust, and each driver will continue to use the driving assistance system.
[0156] <<3. Third Embodiment>> Next, a driving assistance device 50 according to a third embodiment of the present disclosure will be described. In the second embodiment, an example in which the driving assistance device 50 is applied to a situation in which the strength of a traffic rule transitions spatially will be described. In the third embodiment, an example in which the driving assistance device 50 is applied to a situation in which the strength of a traffic rule transitions temporally will be described.
[0157] 33 shows an example of a situation to which this embodiment can be applied. Consider a situation in which a vehicle 1 traveling in lane 113 turns right at an intersection and proceeds to lane 115 after the right turn. The risk at the intersection changes in accordance with the gradual change in the color of the traffic signal 111. Specifically, the risk of violating the rules of the intersection increases in stages as the color of the traffic signal 111 changes in one or more stages from green (first color), which indicates that the vehicle 1 is permitted to pass, to red (second color), which indicates that the vehicle 1 is prohibited from passing.
[0158] The basic method described in the first embodiment does not reflect the concept of time-varying traffic rules. Specifically, when the allowable risk shown in the above formula (2) is used, the target speed is reduced when traveling through an area with a high risk value, or the vehicle 1 stops before entering the area with a high risk value. Therefore, when the color of the traffic signal 111 changes from yellow to red after the vehicle 1 enters the intersection, the vehicle 1 will travel at a low speed or stop within the intersection.
[0159] In this embodiment, the intensity of the traffic rule is set in advance according to the lighting color of the traffic signal 111, and the maximum risk value C i When traffic rules change over time, the speed dependency of the allowable risk of deviation from the rules is changed depending on the positional relationship between the vehicle 1 and the applicable range of the traffic rules, thereby enabling the vehicle 1 to travel in accordance with the traffic rules.
[0160] 34 shows an example of setting the intensity of a traffic rule that changes depending on the lighting color of the traffic signal 111. In the setting example shown in FIG. 34, the intensity of the traffic rule is set to "low" when the lighting color of the traffic signal 111 is green. The intensity of the traffic rule is set to "medium" when the lighting color of the traffic signal 111 is yellow. The intensity of the traffic rule is set to "high" when the lighting color of the traffic signal 111 is red. The intensity of the traffic rule according to the lighting color of the traffic signal 111 corresponds to the level of the rule violation risk set at the intersection.
[0161] 35 to 37 show changes in the risk map when the color of the traffic signal 111 changes over time from green to yellow to red. In the examples shown, the rule deviation risk set at an intersection is set in the area where the intersecting roads overlap. Note that crosswalks are not shown in FIGS. 35 to 37.
[0162] When the traffic signal 111 is lit green, the rule deviation risk set for the intersection is "low" ( FIG. 35 ). When the traffic signal 111 changes to yellow, the rule deviation risk set for the intersection changes to "medium" ( FIG. 36 ). Note that when the traffic signal 111 is lit green and the signal on the pedestrian 117 side begins to flash, the rule deviation risk may be gradually increased. Furthermore, when the traffic signal 111 changes to red, the rule deviation risk set for the intersection changes to "high."
[0163] When the vehicle 1 is approaching an intersection, the allowable risk setting unit 69 sets the allowable road departure risk of the above formula (5) for the rule deviation risk set for the intersection so that whether the vehicle 1 can enter the intersection is determined according to the lighting color of the traffic signal 111. In this way, an allowable risk that is inversely proportional to the speed of the vehicle 1 about to enter the intersection is set, and driving conditions for slowing down or stopping the vehicle 1 are set according to the rule deviation risk set for the intersection according to the lighting color of the traffic signal 111.
[0164] 35, when the vehicle 1 has already entered the intersection, the allowable risk setting unit 69 sets the allowable rule deviation risk of the above formula (4) for the rule deviation risk set for the intersection. As a result, when the rule deviation risk set for the intersection exceeds the allowable rule deviation risk as the color of the traffic signal 111 changes from green to yellow to red, the allowable risk setting unit 69 increases the set speed of the vehicle 1 to increase the allowable rule deviation risk. In other words, it is possible to realize the judgment of a person who tries to quickly escape from the rule deviation state in which the vehicle 1 remains at the intersection even after the color of the traffic signal 111 turns red.
[0165] FIG. 38 shows a flowchart of the process for setting the risk potential and the tolerable risk according to this embodiment.
[0166] First, the risk setting unit 67 sets a rule deviation risk R that is set in advance according to the type of the static risk object present on the road in the same procedure as in step S51 of the flowchart shown in FIG.rul and road departure risk R dev (Step S51) A static risk object refers to a risk object for which traffic rules do not change over time, such as a boundary line with a specified type or a curb indicating a road boundary.
[0167] Next, the risk setting unit 67 determines a rule deviation risk R for dynamic risk objects present on the road based on the current state. rul A dynamic risk object refers to a risk object where traffic rules change over time, such as an intersection where the possibility of entry changes depending on the lighting color of the traffic signal 111 described above. In the example where the vehicle 1 passes through an intersection described above, a rule deviation risk R that is preset according to the current lighting color of the traffic signal 111 is set. rul Set.
[0168] The driving assistance device 50 can recognize the lighting color of the traffic signal 111 using, for example, the forward-facing cameras 23LF, 23RF, or can obtain information about the lighting color of the traffic signal 111 through road-to-vehicle communication or the like.
[0169] Next, the allowable risk setting unit 69 sets an allowable risk for each of the set risk potentials (step S53). At this time, the allowable risk for the rule deviation risk set for the dynamic risk object is set according to the speed of the vehicle 1 and the positional relationship between the risk object and the vehicle 1. In the example where the vehicle 1 passes through an intersection described above, when the vehicle 1 is about to enter the intersection, the allowable road departure risk U ≠ 1 is calculated based on the above formula (5) for the rule deviation risk set for the intersection according to the vehicle speed. dev As a result, the vehicle 1 can approach the intersection as the vehicle speed decreases, and driving conditions are set to decelerate or stop the vehicle 1 according to the lighting color of the traffic signal 111.
[0170] On the other hand, when the vehicle 1 enters an intersection, the rule deviation allowable risk U is calculated based on the above formula (4) in accordance with the vehicle speed for the rule deviation risk set at the intersection. rul This sets the rule deviation risk R rulThe larger the value, the greater the tolerable risk of rule deviation. rul is the risk of rule violation R rul Therefore, a driving condition is set that allows the vehicle 1 to quickly leave the intersection when the color of the traffic signal 111 changes from yellow to red.
[0171] As described above, according to the example of this embodiment, in a situation where the strength of traffic rules changes over time on the travel route of the vehicle 1, it is possible to set driving conditions that enable the vehicle to quickly escape from a rule-violating state as the strength of the traffic rules increases. Therefore, it is possible to realize a driving assistance system that can give drivers a sense of security and trust, and each driver will continue to use the driving assistance system.
[0172] (Other Application Examples of the Present Embodiment) The situations in which the example of the driving assistance process according to the present embodiment can be applied are not limited to the above-described situation in which the vehicle 1 passes through an intersection. For example, the example of the driving assistance process according to the present embodiment can also be applied to a situation in which the vehicle 1 is traveling in a specific vehicle priority lane in which a time period during which traffic rules apply is set.
[0173] 39 to 41 show a scene in which a vehicle 1 traveling in a lane adjacent to a bus-only lane for which an applicable time period has been set overtakes another vehicle 95 traveling in front. rul is set, and the road departure risk R dev In addition, the risk of deviation from the bus lane rule R is calculated according to the margin of time from the current time until the bus lane application time. rul Each diagram shows the set risk of rule deviation R rul and road departure risk R dev The risk map shows the risk of deviation from the rules in a cross section perpendicular to the route. rul and road departure risk R dev is shown.
[0174] The applicable time period for bus lanes is from 7:00 AM to 9:00 AM, and Figures 39 to 41 show risk maps set five minutes before the start time of the applicable time period for bus lanes (6:55 AM), two minutes before the start time (6:58 AM), and the start time (7:00 AM), respectively. The driving assistance device 50 can recognize markings printed on the road surface, for example, using the forward-facing cameras 23LF and 23RF. Alternatively, the driving assistance device 50 may obtain information about bus lanes and their applicable time periods based on the current position of the vehicle 1 and map data, or may obtain information about bus lanes and their applicable time periods by communicating with an external information processing device.
[0175] If there is sufficient time between the current time and the applicable time zone of the bus lane, the entry of the vehicle 1 into the bus lane is not restricted, and the tolerance for crossing the boundary line 107 is high. Therefore, as shown in FIG. 39 , the rule deviation risk set for the bus lane is zero, and the rule deviation risk R rul In this situation, the risk potential of the vehicle lane and the bus lane are the same, so the tolerable risk setting unit 69 sets the rule deviation risk R rul and road departure risk R dev The learned rule deviation tolerance risk U rul and road departure allowable risk U dev This sets the allowable risk of rule deviation U rul and road departure allowable risk U dev is the risk of rule violation R rul and road departure risk R dev The travel path of the vehicle 1 is set to a position close to the reference path within the range exceeding each of the above, and the vehicle 1 can overtake the other vehicle 95.
[0176] As the time when the bus lane is to be activated approaches, the rule deviation risk R rul While maintaining the same, the bus lane has a positive rule violation risk R rulFurthermore, when the time comes when the bus lane is valid, a rule deviation risk R is set on the boundary line 107 as shown in FIG. 41 in order to restrict the vehicle 1 from entering the bus lane. rul , and the rule deviation risk R set for the bus-only lane rul In other words, the closer the current time is to the applicable time zone of the bus-only lane, the greater the rule deviation risk R rul Increase the value.
[0177] In this situation, when the vehicle 1 is traveling in the own lane, the tolerable risk setting unit 69 sets the rule deviation risk R rul and the road departure risk R dev The learned rule deviation tolerance risk U rul and road departure allowable risk U dev This sets the allowable risk of rule deviation U rul and road departure allowable risk U dev is the risk of rule violation R rul and road departure risk R dev The travel path of the vehicle 1 is set to a position close to the reference path within the range exceeding each of the above.
[0178] On the other hand, while the vehicle 1 is traveling in the bus travel lane, the tolerable risk setting unit 69 sets the rule deviation risk R rul and the road departure risk R dev The learned rule deviation tolerance risk U rul and road departure allowable risk U dev The allowable risk setting unit 69 also sets the rule deviation risk R rul Based on the above formula (4), the rule deviation allowable risk U rul As a result, the risk of deviation from the rule at the traveling position of the vehicle 1 is set as R rul is the allowable risk of deviation from the rules. rul If the speed exceeds the allowable risk of deviation from the rule, the allowable risk setting unit 69 increases the set speed of the vehicle 1 to set the allowable risk of deviation from the rule U rulIn other words, it is possible to realize the judgment of a person who wants to quickly escape from the state of violation of the rules, that is, the state in which the vehicle 1 is traveling in the bus lane.
[0179] Although the preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings, the present invention is not limited to these examples. It is clear that a person skilled in the art to which the present invention pertains can conceive of various modifications and alterations within the scope of the technical ideas set forth in the claims, and it is understood that these also naturally fall within the technical scope of the present invention.
[0180] For example, in the above embodiment, an example in which the driving assistance device 50 is configured as a single device has been described, but the technology of the present disclosure is not limited to such an example. Some or all of the functions of the driving assistance device 50 may be provided in an external server installed outside the vehicle.
[0181] In addition, the technology of the present disclosure can also be realized as a vehicle equipped with the driving assistance device described in the above embodiment, a driving assistance processing method using the driving assistance device, a computer program that causes a computer to function as the above driving assistance device, and a non-transitory tangible recording medium on which the computer program is recorded.
[0182] 1: Vehicle 50: Driving assistance device 51: Processing unit 53: Storage unit 55: Learning data storage unit 61: Acquisition unit 63: Surrounding environment detection unit 65: Learning processing unit 67: Risk setting unit 69: Tolerable risk setting unit 71: Driving condition setting unit 73: Notification processing unit R col : Collision risk R dev : Road departure risk R rul :Risk of deviation from rules U col : Collision tolerance risk U dev : Allowable risk of road departure U rul : Tolerable risk of deviation from rules
Claims
1. A driving assistance device that assists in driving a vehicle, comprising one or more processors and one or more memories communicatively connected to the one or more processors, wherein the one or more processors execute the following operations: an acquisition process that acquires information about the environment around the vehicle; a risk setting process that sets a rule deviation risk, which indicates the risk of deviating from a traffic rule related to driving of the vehicle, for a target of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule is applied; an allowable risk setting process that sets an allowable rule deviation risk, which indicates the allowable range of driving of the vehicle against the rule deviation risk, so that it increases in proportion to the speed of the vehicle; and a driving condition setting process that sets a target trajectory and target speed during automatic driving of the vehicle based on the rule deviation risk and the allowable rule deviation risk so that the vehicle drives at a position where the allowable rule deviation risk exceeds the rule deviation risk.
2. The driving assistance device of claim 1, wherein the one or more processors set the risk of violating a rule as the strength of the traffic rule depending on at least one of the type of boundary line between a first lane in which the vehicle is traveling and a second lane adjacent to the first lane, or the type of a restricted driving zone set on the road.
3. The driving assistance device of claim 2, wherein the type of boundary line varies depending on the tolerance for the vehicle crossing the boundary line, and the one or more processors increase the risk of rule violation as the type of boundary line becomes lower in tolerance for the vehicle crossing the boundary line.
4. The driving assistance device of claim 2, wherein the driving restriction zones are of different types depending on the tolerance for the vehicle entering the driving restriction zone, and the one or more processors increase the risk of deviation from the rules as the type of driving restriction zone increases the tolerance for the vehicle entering the driving restriction zone.
5. The driving assistance device according to claim 1, wherein the one or more processors set the risk of violating a rule as the strength of the traffic rule according to the color of a traffic light.
6. A driving assistance device as described in claim 5, wherein the color of the traffic light changes in one or more steps from a first color indicating that the vehicle is permitted to pass to a second color indicating that the vehicle is prohibited from passing, and the one or more processors change the color of the traffic light so that the risk of violating the rules increases each time the color of the traffic light changes from the first color to the second color.
7. The driving assistance device according to claim 1, wherein the one or more processors set the risk of deviation from a rule according to the applicable time period of a specific vehicle priority lane in which the applicable time period of the traffic rule is set and the current time.
8. The driving assistance device according to claim 7, wherein the one or more processors increase the risk of deviation from the rule the closer the current time is to the applicable time zone.
9. The driving assistance device of claim 1, wherein the one or more processors, in the risk setting process, further set a collision risk indicating the risk of collision with an obstacle; in the allowable risk setting process, set a tolerable collision risk indicating an allowable range of driving of the vehicle against the collision risk so as to decrease in proportion to the vehicle speed; and in the driving condition setting process, set a target trajectory and target speed for the vehicle during automatic driving based on the rule deviation risk and the collision risk, the rule deviation allowable risk and the collision allowable risk, so that the vehicle drives at a position where the rule deviation allowable risk exceeds the rule deviation risk and where the collision allowable risk exceeds the collision risk.
10. The driving assistance device of claim 1, wherein the one or more processors, in the risk setting process, further set a road departure risk indicating the risk of going off a road, and in the acceptable risk setting process, set a road departure tolerable risk indicating an acceptable driving range of the vehicle against the road departure risk so as to decrease in proportion to the vehicle speed, and in the driving condition setting process, set a target trajectory and a target speed for the vehicle during automatic driving based on the rule deviation risk and the road departure risk, and the rule deviation tolerable risk and the road departure tolerable risk, so that the vehicle drives at a position where the rule deviation tolerable risk exceeds the rule deviation risk and where the road departure tolerable risk exceeds the road departure tolerable risk.
11. A driving assistance processing method in which a computer performs the following steps: acquires information about the environment surrounding a vehicle; sets a rule deviation risk, indicating the risk of deviating from a traffic rule regarding the driving of the vehicle, for the subject of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule applies; sets a rule deviation allowable risk, indicating the tolerable range of driving of the vehicle against the rule deviation risk, to increase in proportion to the speed of the vehicle; and sets a target trajectory and target speed for the vehicle during automatic driving based on the rule deviation risk and the rule deviation allowable risk so that the vehicle drives at a position where the rule deviation allowable risk exceeds the rule deviation risk.
12. A non-transitory tangible recording medium having recorded thereon a computer program that causes a computer to perform the following operations: acquire information about the environment surrounding the vehicle; set a rule deviation risk, indicating the risk of deviating from a traffic rule regarding the driving of the vehicle, for the subject of the traffic rule in accordance with at least one of the strength of the traffic rule or the time period during which the traffic rule applies; set a rule deviation allowable risk, indicating the tolerable range of driving of the vehicle relative to the rule deviation risk, to increase in proportion to the speed of the vehicle; and set a target trajectory and target speed for the vehicle during autonomous driving based on the rule deviation risk and the rule deviation allowable risk so that the vehicle drives at a position where the rule deviation allowable risk exceeds the rule deviation risk.
Citation Information
Patent Citations
Driving support method and driving support device
JP2021196874A
Driving support control device
JP2022054279A
Driving support device
JP2022129190A
