Driver Assistance Systems

Abstract

The present invention relates to a driving assistance system. A risk potential field used for risk avoidance control (deceleration control) includes an obstacle potential field. The obstacle potential field is a risk potential field in which the risk value is maximum at the location of an object and decreases as the distance from the object increases. The search range is the range between the location of the object and a location that is a predetermined gap away from the object. A minimum point (valley) of the risk potential field is searched within the search range. Deceleration control is performed based on the positional relationship between the minimum point and the object. If multiple minimum points exist in the search range, the minimum point closest to the object is selected.

Description

Technical Field

[0001] The present invention relates to a driving assistance control system for assisting driving of a vehicle, and more particularly to a risk avoidance control system for reducing the risk of collision with an object in front of the vehicle. Background Art

[0002] Japanese Patent Application Laid-Open No. 2018-203034 discloses a driving route determination device for determining a vehicle's driving route. The driving route determination device uses a risk potential area and a benefit potential area to determine the vehicle's driving route. The risk potential area indicates areas where obstacles such as pedestrians and other vehicles may be present. The benefit potential area indicates the ideal driving area for the vehicle. This benefit potential area is set based on driving data from experienced drivers.

[0003] We will consider "risk avoidance control" to reduce the risk of collision with an object ahead of the vehicle. In particular, we will consider risk avoidance control based on a risk potential field. The risk potential field represents the risk associated with vehicle travel as a function of position.

[0004] Risk avoidance control includes the following deceleration control. First, a minimum point (valley) in the risk potential field is searched for within a search range near a target object. Deceleration control is performed based on the positional relationship between this minimum point and the target object. Multiple minimum points may be found within the search range. If an inappropriate minimum point is selected from these multiple minimum points, the positional relationship between the minimum point and the target object also becomes inappropriate, resulting in inappropriate deceleration control. Vehicle occupants may experience discomfort from inappropriate risk avoidance control. Summary of the Invention

[0005] One of the objects of the present invention is to provide a technology capable of suppressing discomfort with risk avoidance control based on a risk potential field.

[0006] One aspect of the present invention relates to a driving assistance system for assisting driving of a vehicle.

[0007] The driving assistance system includes: a storage device storing driving environment information indicating a driving environment of a vehicle; and a processor executing risk avoidance control for reducing a risk of collision with an object in front of the vehicle based on the driving environment information.

[0008] The risk potential field shows the risk value as a function of position.

[0009] The obstacle potential field is a risk potential field in which the risk value is the largest at the location of the target object and decreases as the distance from the target object increases.

[0010] The risk avoidance control includes a deceleration control for decelerating the vehicle.

[0011] The risk potential field used for deceleration control includes at least an obstacle potential field. The search range is a range between the position of the target object and a position separated from the target object by a predetermined gap.

[0012] The processor is configured to: set a risk potential field for deceleration control based on driving environment information; search for a minimum point of the risk potential field for deceleration control from a search range; and perform deceleration control based on a positional relationship between the minimum point and an object.

[0013] When there are multiple minimum points in the search range, the processor selects the minimum point closest to the target object among the multiple minimum points as the minimum point.

[0014] According to the present invention, a risk potential field including an obstacle potential field is applied to deceleration control of risk avoidance control. The obstacle potential field is a risk potential field in which the risk value is maximum at the position of the target object and decreases as the distance from the target object increases. The search range is the range between the position of the target object and the position separated from the target object by a predetermined gap. The minimum point (valley) of the risk potential field is searched in the search range. Then, deceleration control is performed based on the positional relationship between the minimum point and the target object. When there are multiple minimum points in the search range, the minimum point closest to the target object is selected. As a result, an appropriate positional relationship between the minimum point and the target object is obtained, and deceleration control is appropriately performed. That is, it is possible to suppress discomfort and uneasiness with risk avoidance control (deceleration control) based on the risk potential field. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Hereinafter, features, advantages, and technical and industrial significance of exemplary embodiments of the present invention will be described with reference to the accompanying drawings, wherein like reference numerals denote like elements, and wherein:

[0016] Figure 1 This is a conceptual diagram for explaining the outline of the driving assistance system according to the embodiment of the present invention.

[0017] Figure 2 This is a conceptual diagram for explaining an example of risk avoidance control according to an embodiment of the present invention.

[0018] Figure 3 This is a block diagram showing an example configuration of a vehicle and a driving assistance system according to an embodiment of the present invention.

[0019] Figure 4 This is a block diagram showing an example of driving environment information in the embodiment of the present invention.

[0020] Figure 5 This is a conceptual diagram for explaining the obstacle potential field according to the embodiment of the present invention.

[0021] Figure 6This is a conceptual diagram for explaining the vehicle central potential field according to the embodiment of the present invention.

[0022] Figure 7 This is a conceptual diagram for explaining the lane center potential field used in the comparative example.

[0023] Figure 8 This is a conceptual diagram for explaining the outline of steering control based on a risk potential field.

[0024] Figure 9 This is a block diagram for explaining an outline of steering control according to an embodiment of the present invention.

[0025] Figure 10 This is a conceptual diagram for explaining an example of steering control according to the embodiment of the present invention.

[0026] Figure 11 This is a flowchart showing processing related to steering control according to the embodiment of the present invention.

[0027] Figure 12 Yes Figure 11 Flowchart of the processing example in step S120 in .

[0028] Figure 13 This is a conceptual diagram for explaining the margin time to reach a target object.

[0029] Figure 14 This is a conceptual diagram for explaining steering control according to the embodiment of the present invention.

[0030] Figure 15 This is a conceptual diagram for explaining steering control according to the embodiment of the present invention.

[0031] Figure 16 This is a conceptual diagram for explaining an example of a plurality of first minimum point candidates according to the embodiment of the present invention.

[0032] Figure 17 This is a conceptual diagram for explaining a method of selecting an appropriate first minimum point from a plurality of first minimum point candidates according to an embodiment of the present invention.

[0033] Figure 18 This is a conceptual diagram for explaining the outline of deceleration control according to the embodiment of the present invention.

[0034] Figure 19 This is a conceptual diagram for explaining the suppression amount used in the deceleration control according to the embodiment of the present invention.

[0035] Figure 20 This is a block diagram for explaining the outline of deceleration control according to the embodiment of the present invention.

[0036] Figure 21 This is a conceptual diagram for explaining an example of deceleration control according to the embodiment of the present invention.

[0037] Figure 22 This is a flowchart showing processing related to deceleration control according to the embodiment of the present invention.

[0038] Figure 23 Yes Figure 22 Flowchart of the processing example in step S220 in .

[0039] Figure 24 This is a conceptual diagram for explaining deceleration control according to the embodiment of the present invention.

[0040] Figure 25 This is a conceptual diagram for explaining a method of selecting an appropriate minimum point from a plurality of minimum point candidates according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] Embodiments of the present invention will be described with reference to the accompanying drawings.

[0042] 1. Driver Assistance Systems

[0043] 1-1. Overview

[0044] Figure 1 This is a conceptual diagram used to outline the driving assistance system 10 of this embodiment. The driving assistance system 10 performs "driving assistance control" to assist in driving the vehicle 1. Driving assistance control can also be included in autonomous driving control. Typically, the driving assistance system 10 is installed in the vehicle 1. Alternatively, at least a portion of the driving assistance system 10 may be located in an external device outside the vehicle 1, and driving assistance control may be performed remotely. In other words, the driving assistance system 10 may be distributed between the vehicle 1 and an external device.

[0045] Driving assistance control includes "risk avoidance control" to proactively avoid risks ahead of vehicle 1. More specifically, driving assistance system 10 identifies object 5 ahead of vehicle 1. Then, driving assistance system 10 executes risk avoidance control to proactively reduce (avoid) the risk of collision with object 5. This risk avoidance control includes at least one of steering control to steer vehicle 1 and deceleration control to slow vehicle 1.

[0046] For example, in Figure 1In the example, vehicle 1 is traveling in lane LA in roadway 2. There is a pedestrian 5A in the roadside area 3 (road shoulder, roadside strip, sidewalk, etc.) adjacent to roadway 2. Pedestrian 5A may enter lane LA. Therefore, pedestrian 5A in roadside area 3 is a risk to vehicle 1. In order to reduce the risk of collision with pedestrian 5A, driving assistance system 10 performs risk avoidance control as needed. For example, driving assistance system 10 automatically steers vehicle 1 away from pedestrian 5A. Figure 1 In FIG, a trajectory TR0 shows the trajectory of the vehicle 1 when the risk avoidance control is not executed. On the other hand, a trajectory TR1 shows the trajectory of the vehicle 1 when the risk avoidance control is executed.

[0047] The pedestrian 5A may be replaced by a bicycle or a two-wheeled vehicle. In addition, not only the roadside area 3 but also pedestrians, bicycles, two-wheeled vehicles, and preceding vehicles present in the roadway 2 become targets of risk avoidance control.

[0048] Figure 2 This is a conceptual diagram for explaining another example of risk avoidance control. The target of risk avoidance control is not limited to the "manifest risk" such as the pedestrian 5A mentioned above, but may also include "potential risk". For example, Figure 2 In the example, there is a parked vehicle 5B in the roadside area 3 in front of vehicle 1. The area in front of parked vehicle 5B is a blind spot, and pedestrian 5C could potentially escape from this blind spot. Therefore, parked vehicle 5B in front of vehicle 1 presents a risk to vehicle 1 and is the target of risk avoidance control. For example, driving assistance system 10 automatically steers vehicle 1 away from parked vehicle 5B.

[0049] As described above, the object 5 that is the target of the risk avoidance control includes at least one of a pedestrian, a bicycle, a two-wheeled vehicle, and another vehicle in front of the vehicle 1 .

[0050] Here, the coordinate system and direction are defined. The vehicle coordinate system (X, Y) is a relative coordinate system fixed to the vehicle 1 and changes as the vehicle 1 moves. The X direction is the forward direction (travel direction) of the vehicle 1. The Y direction is the lateral direction of the vehicle 1. The X and Y directions are mutually orthogonal. The LX direction (lane length dimension) is the direction in which the lane LA extends. The LY direction (lane width dimension) is the width direction of the lane LA. The LX and LY directions are mutually orthogonal. The longitudinal distance is the distance in the X or LX direction. The lateral distance is the distance in the Y or LY direction.

[0051] 1-2. Example of structure

[0052] Figure 3 1 is a block diagram schematically showing an example of the configuration of the vehicle 1 and the driving assistance system 10 according to the present embodiment. Figure 3A configuration example related to risk avoidance control is shown. The vehicle 1 includes a sensor group 20 and a travel device 30 .

[0053] The sensor group 20 includes a position sensor 21, a vehicle state sensor 22, and an identification sensor 23. The position sensor 21 detects the position and orientation of the vehicle 1 in the absolute coordinate system. As the position sensor 21, a GPS (Global Positioning System) sensor is exemplified. The vehicle state sensor 22 detects the state of the vehicle 1. As the vehicle state sensor 22, a vehicle speed sensor, a yaw rate sensor, a lateral acceleration sensor, a steering angle sensor, etc. are exemplified. The identification sensor 23 identifies (detects) the conditions around the vehicle 1. As the identification sensor 23, a camera, a radar, a laser radar (LIDAR: Laser Imaging Detection and Ranging) and the like are exemplified.

[0054] The travel device 30 includes a steering device 31, a drive device 32, and a brake device 33. The steering device 31 steers the wheels of the vehicle 1. For example, the steering device 31 includes an electric power steering (EPS) device. The drive device 32 is a power source that generates driving force. Examples of the drive device 32 include an engine, an electric motor, and an in-wheel motor. The brake device 33 generates braking force.

[0055] The driving assistance system 10 includes at least a control device 100 . The driving assistance system 10 may also include a sensor group 20 . The driving assistance system 10 may also include a travel device 30 .

[0056] The control device 100 controls the vehicle 1. Typically, the control device 100 is a microcomputer mounted on the vehicle 1. The control device 100 is also called an ECU (Electronic Control Unit). Alternatively, the control device 100 may be an information processing device external to the vehicle 1. In this case, the control device 100 communicates with the vehicle 1 and remotely controls the vehicle 1.

[0057] The control device 100 includes a processor 110 and a storage device 120. The processor 110 executes various processes. The storage device 120 stores various information. Examples of the storage device 120 include volatile memory and nonvolatile memory. The processor 110 executes a control program, which is a computer program, thereby implementing the various processes performed by the processor 110 (control device 100). The control program is stored in the storage device 120 or recorded on a computer-readable recording medium.

[0058] 1-3. Information Acquisition Processing

[0059] Processor 110 (control device 100 ) executes “information acquisition processing” to acquire driving environment information 200 indicating the driving environment of vehicle 1 . Driving environment information 200 is acquired based on detection results from sensor group 20 mounted on vehicle 1 . Acquired driving environment information 200 is stored in storage device 120 .

[0060] Figure 4 2 is a block diagram showing an example of driving environment information 200. Driving environment information 200 includes vehicle position information 210, vehicle state information 220, surrounding situation information 230, map information 260, and the like.

[0061] The vehicle position information 210 is information indicating the position and orientation of the vehicle 1 in the absolute coordinate system. The processor 110 acquires the vehicle position information 210 based on the detection result of the position sensor 21 .

[0062] The vehicle state information 220 is information indicating the state of the vehicle 1 . Examples of the state of the vehicle 1 include vehicle speed, yaw rate, lateral acceleration, and steering angle. The processor 110 acquires the vehicle state information 220 based on the detection results of the vehicle state sensor 22 .

[0063] Surrounding situation information 230 is information indicating the conditions surrounding vehicle 1. Surrounding situation information 230 includes information obtained by recognition sensor 23. For example, surrounding situation information 230 includes image information captured by a camera and indicating the conditions surrounding vehicle 1. As another example, surrounding situation information 230 includes measurement information obtained by radar or lidar. Furthermore, surrounding situation information 230 includes road configuration information 240 and landmark information 250.

[0064] Road configuration information 240 is information related to the road configuration surrounding vehicle 1. The road configuration surrounding vehicle 1 includes demarcation lines (white lines) and road edge objects. Road edge objects are three-dimensional obstacles that indicate the edge of the road. Examples of road edge objects include curbs, guardrails, walls, and medians. Road configuration information 240 indicates at least the positions of demarcation lines and road edge objects (their relative positions with respect to vehicle 1).

[0065] For example, by analyzing image information captured by a camera, demarcation lines can be identified and their relative positions calculated. Examples of image analysis methods include semantic segmentation and edge detection. Similarly, by analyzing image information, objects at the road's edge can be identified and their relative positions calculated. Alternatively, the relative positions of objects at the road's edge can be obtained based on radar measurement information.

[0066] Object information 250 is information related to objects 5 surrounding vehicle 1. Examples of objects 5 include pedestrians, bicycles, two-wheeled vehicles, and other vehicles (preceding vehicles, parked vehicles). Object information 250 indicates the relative position and relative speed of the objects relative to vehicle 1. For example, by analyzing image information obtained by a camera, it is possible to identify objects 5 and calculate their relative positions. Alternatively, it is possible to identify objects 5 based on radar measurement information and obtain their relative positions and relative speeds. Object information 250 may include the movement direction and movement speed of object 5. The movement direction and movement speed of object 5 can be calculated by tracking the position of object 5. Object information 250 may indicate the type of object 5 (pedestrian, bicycle, two-wheeled vehicle, other vehicle, etc.).

[0067] Map information 260 indicates lane configuration, road shape, and other information. The control device 100 obtains map information 260 for the required area from a map database. The map database may be stored in a predetermined storage device mounted on the vehicle 1 or in a management server external to the vehicle 1. In the latter case, the processor 110 communicates with the management server to obtain the required map information 260.

[0068] 1-4. Vehicle driving control

[0069] Processor 110 (control device 100) performs "vehicle driving control" to control the driving of vehicle 1. Vehicle driving control includes steering control to control the steering of vehicle 1, acceleration control to control the acceleration of vehicle 1, and deceleration control to control the deceleration of vehicle 1. Processor 110 performs vehicle driving control by controlling driving device 30. Specifically, processor 110 performs steering control by controlling steering device 31. In addition, processor 110 performs acceleration control by controlling driving device 32. In addition, control device 100 performs deceleration control by controlling braking device 33.

[0070] 1-5. Risk avoidance control

[0071] Processor 110 (control device 100) executes driving assistance control to assist in driving vehicle 1. Driving assistance control includes risk avoidance control. Risk avoidance control is vehicle driving control designed to reduce (avoid) the risk of collision with an object 5 in front of vehicle 1 and includes at least one of steering control and deceleration control. Processor 110 executes risk avoidance control based on the aforementioned driving environment information 200.

[0072] The risk avoidance control according to this embodiment will be described in more detail below.

[0073] 2. Risk potential field

[0074] As a value representing the risk associated with vehicle travel, a "risk value R (risk potential)" is introduced. The risk value R is defined for each location. Locations with high risk values ​​R are locations that vehicle 1 should avoid. The "risk potential field U" represents the risk value R as a function of location. In other words, the risk potential field U represents the distribution of risk values ​​R.

[0075] It should be noted that "position" can be a position in the vehicle coordinate system (X, Y) or a position in the absolute coordinate system (latitude, longitude). Coordinate transformation between the absolute coordinate system and the vehicle coordinate system can be performed based on vehicle position information 210. In the following description, the position in the vehicle coordinate system and the position in the absolute coordinate system are considered equivalent.

[0076] The risk avoidance control (steering control, deceleration control) of the present embodiment is executed based on the risk potential field U. The components of the risk potential field U will be described below.

[0077] 2-1. Obstacle Potential Field

[0078] Figure 5 This is a conceptual diagram for explaining the obstacle potential field Uo. The obstacle potential field Uo is a risk potential field U for preventing the vehicle 1 from approaching the target object 5. Therefore, the risk value R shown in the obstacle potential field Uo is the largest at the target object 5 and decreases as the vehicle moves away from the target object 5.

[0079] More specifically, the obstacle potential field Uo represents a two-dimensional distribution of risk values ​​R. Figure 5 The following diagram shows the distribution profiles along two principal axis directions. The two principal axis directions are the LX direction (lane length direction) and the LY direction (lane width direction). As another example, the two principal axis directions may be the X and Y directions. Object position PT is the location of object 5. In each principal axis direction, the risk value R is greatest at object position PT and decreases with distance from object position PT. In other words, the distribution of risk value R has a mountainous shape.

[0080] The obstacle potential function fo is a distribution function that represents the distribution of the risk value R for the obstacle potential field Uo. For example, the obstacle potential function fo is a Gaussian function. In this case, the distribution is Gaussian (normal). The distribution parameters σx and σy represent the degree of spread of the distribution in the two principal axis directions. When the distribution is Gaussian, the distribution parameters σx and σy represent the standard deviation.

[0081] The distribution parameters σx and σy may differ depending on the type of the object 5. For example, the distribution parameters σx and σy when the object 5 is a pedestrian are larger than when the object 5 is another vehicle.

[0082] The distribution parameters σx and σy may also vary according to the vehicle speed of the vehicle 1. For example, as the vehicle speed increases, the distribution parameters σx and σy increase. In this case, the distribution parameters σx and σy are given by a map.

[0083] Potential function information 300 (see Figure 3 ) represents the obstacle potential function fo and distribution parameters σx, σy. The potential function information 300 is generated in advance and stored in the storage device 120.

[0084] Processor 110 sets an obstacle potential field Uo associated with object 5. The location and type of object 5 are obtained from object information 250. The configuration of lane LA is obtained from road configuration information 240 or map information 260. The LX and LY directions are obtained from the configuration of lane LA. The vehicle speed is obtained from vehicle state information 220. Therefore, processor 110 can set obstacle potential field Uo associated with object 5 based on driving environment information 200 and potential function information 300.

[0085] 2-2. Vehicle Center Potential Field

[0086] Figure 6 This is a conceptual diagram illustrating the vehicle center potential field Ue. The lane LA in which vehicle 1 is located is the area sandwiched between the left and right lane boundaries LB (dividing lines). Lane LA and lane boundaries LB extend in the LX direction (the length of the lane). The vehicle center potential field Ue represents the risk potential field U for driving vehicle 1 along lane LA. Therefore, the "valley Ve" of the risk value R shown in the vehicle center potential field Ue extends in the LX direction.

[0087] More specifically, the vehicle center potential field Ue represents the two-dimensional distribution of the risk value R. Figure 6 The following figure shows the profile of the distribution along the LY direction (lane width direction). The vehicle lateral position PV is the position of the vehicle 1 in the LY direction. In the LY direction, the risk value R is minimum at the vehicle lateral position PV and increases as it moves away from the vehicle lateral position PV. In other words, the risk value distribution has a U-shaped shape. The position of the valley Ve of the risk value R coincides with the vehicle lateral position PV. The valley Ve extends from the position of the vehicle 1 in the LX direction. In other words, the position of the valley Ve is not fixed but changes dynamically in conjunction with the position of the vehicle 1.

[0088] The vehicle center potential function fe is a distribution function that represents the distribution of the risk value R of the vehicle center potential field Ue. For example, the vehicle center potential function fe is a quadratic curve. The distribution parameter σe is a parameter that represents the degree of spread of the distribution. Potential function information 300 (see Figure 3 ) further represents the vehicle central potential function fe and distribution parameter σe.

[0089] Processor 110 sets the vehicle center potential field Ue. The position of vehicle 1 is obtained based on vehicle position information 210. The configuration of lane LA is obtained based on road configuration information 240 or map information 260. The LX direction and LY direction are obtained based on the configuration of lane LA. Therefore, processor 110 can set the vehicle center potential field Ue based on driving environment information 200 and potential function information 300.

[0090] 2-3. Lane Center Potential Field

[0091] Figure 7 represents the lane center potential field Ur. The lane center potential field Ur is the risk potential field U for causing the vehicle 1 to travel along the lane center LC. The "valley Vr" of the risk value R indicated by the lane center potential field Ur also extends in the LX direction. However, the position of this valley Vr is fixed at the lane center position PLC (the position of the lane center LC). In other words, the position of the valley Vr of the lane center potential field Ur is fixed to the lane LA and does not change dynamically.

[0092] 3. Steering control based on risk potential field

[0093] 3-1. Overview of Steering Control

[0094] Figure 8 This is a conceptual diagram for explaining the outline of steering control based on the risk potential field U. The overall risk potential field U is obtained by superimposing (adding) the components of the risk potential field U described above. When there are multiple object markers 5, the obstacle potential fields Uo set for each object marker 5 are superimposed.

[0095] In the risk potential field U, there is a “valley” of risk value R. Figure 8 As shown, the valley of the risk potential field U is positioned so as to extend in the LX direction while avoiding the target object 5. By performing steering control so that the vehicle 1 follows the valley of the risk potential field U, the vehicle 1 can be driven while reducing the risk of collision with the target object 5. In other words, risk avoidance control is implemented.

[0096] 3-2. Steering Control Based on the First Risk Potential Field

[0097] Figure 9 This is a block diagram for explaining the outline of the steering control of this embodiment. The first risk potential field U1 is a risk potential field used for steering control. The first risk potential field U1 is the sum of the longitudinal potential field Ux and the obstacle potential field Uo[i], and is represented by the following equation (1).

[0098] [Formula 1]

[0099]

[0100] The obstacle potential field Uo[i] is the obstacle potential field Uo (i=1 to n) associated with the object 5[i]. Here, n is the total number of objects 5 that are the subject of risk avoidance control and is an integer greater than 1. The longitudinal potential field Ux is the vehicle center potential field Ue (see Figure 6 ) or the lane center potential field Ur (refer to Figure 7 ). Preferably, the longitudinal potential field Ux is the vehicle center potential field Ue.

[0101] The first valley V1 is a valley of the risk value R shown in the first risk potential field U1 . The processor 110 performs steering control so that the vehicle 1 follows the first valley V1 .

[0102] Figure 10 This figure illustrates an example of steering control according to this embodiment. A first valley V1 extends from the position of vehicle 1 in the LX direction and then shifts away from object 5. Vehicle 1 initially travels in the LX direction and then turns away from object 5 (trajectory TR1). As the lateral position of vehicle 1 changes, the lateral position of first valley V1 also changes in tandem. Then, first valley V1 extends in the LX direction, and vehicle 1 travels in the LX direction. Vehicle 1 passes to the side of object 5 at an appropriate lateral distance Dy.

[0103] It should be noted that when the longitudinal potential field Ux is the lane center potential field Ur, this lane center potential field Ur constantly generates a force that pulls the vehicle 1 toward the lane center LC. While this force pulling the vehicle 1 toward the lane center LC is preferable for preventing lane departure, it is inherently unrelated to object avoidance. The lane center potential field Ur generates vehicle behavior unrelated to object avoidance, potentially leading to unnecessary or excessive steering control as part of risk avoidance control.

[0104] In this sense, the longitudinal potential field Ux is preferably the vehicle center potential field Ue. The position of the valley Ve of the vehicle center potential field Ue is not fixed but changes dynamically in conjunction with the position of the vehicle 1. Because this valley Ve is reflected in the first valley V1, unnecessary or excessive steering control is suppressed. Suppressing unnecessary or excessive steering control means achieving appropriate vehicle behavior to avoid risks. Consequently, the discomfort felt by the occupants of the vehicle 1 is minimized.

[0105] 3-3. Processing Flow

[0106] Figure 11 This is a flowchart showing the processing related to the steering control in this embodiment. Figure 11 The processing flow shown is repeatedly executed at a fixed cycle.

[0107] 3-3-1. Step S110

[0108] In step S110 , the processor 110 executes the aforementioned information acquisition process. That is, the processor 110 acquires the driving environment information 200 based on the detection results of the sensor group 20 . The driving environment information 200 is stored in the storage device 120 .

[0109] 3-3-2. Step S120

[0110] In step S120, the processor 110 sets the first risk potential field U1. The first risk potential field U1 is the sum of the longitudinal potential field Ux and the obstacle potential field Uo[i] (see formula (1)). The longitudinal potential field Ux is, for example, the vehicle center potential field Ue. The processor 110 sets the vehicle center potential field Ue based on the driving environment information 200 and the potential function information 300. In addition, the processor 110 sets the obstacle potential field Uo[i] for each object mark 5[i] based on the driving environment information 200 and the potential function information 300. Then, the processor 110 sets the sum of the vehicle center potential field Ue and the obstacle potential field Uo[i] as the first risk potential field U1.

[0111] Figure 12 This is a flowchart showing an example of the processing in step S120.

[0112] In step S121, the processor 110 sets the vehicle center potential field Ue based on the driving environment information 200 and the potential function information 300. Then, the processor 110 adds the vehicle center potential field Ue to the first risk potential field U1.

[0113] In step S122, processor 110 determines whether object marker 5 exists in front of vehicle 1 based on object marker information 250. In other words, processor 110 determines whether object marker 5 is recognized in the area in front of vehicle 1. If object marker 5 is recognized in front of vehicle 1 (step S122: Yes), processing proceeds to step S123. Otherwise (step S122: No), step S120 ends.

[0114] In step S123 , the processor 110 determines whether the remaining time T to the recognized object 5 is less than a first time threshold Tth1 .

[0115] Reference Figure 13, the margin time T will be described. Trajectory TR0 shows the trajectory of vehicle 1 when risk avoidance control is not being executed. It is assumed that vehicle 1 is traveling in the LX direction at its current speed. The margin time T is the time until vehicle 1 approaches object 5 under this assumption. Typically, vehicle 1 approaches object 5 when it passes from the side of object 5. The current speed of vehicle 1 is obtained from vehicle state information 220. The position of object 5 is obtained from object information 250. The configuration of lane LA and the LX direction are obtained from road configuration information 240 or map information 260. Therefore, processor 110 can calculate margin time T based on driving environment information 200.

[0116] If the margin time T is less than the first time threshold value Tth1 (step S123 returns Yes), the process proceeds to step S124. Otherwise (step S123 returns No), step S120 ends.

[0117] In step S124, processor 110 sets an obstacle potential field Uo associated with the identified object 5 based on driving environment information 200 and potential function information 300. Processor 110 then adds this obstacle potential field Uo to the first risk potential field U1. Thus, when vehicle 1 approaches object 5 to a certain extent, obstacle potential field Uo associated with this object 5 is added to the first risk potential field U1.

[0118] 3-3-3. Step S130

[0119] In step S130 , the processor 110 sets a forward gaze point PA at a position in front of the vehicle 1 .

[0120] Reference Figure 14 , the forward gaze point PA is described. The forward gaze point PA is set at a first distance S in front of the vehicle 1 along the vehicle's travel direction (X direction). The vehicle's travel direction is determined based on the vehicle position information 210. The first distance S is a fixed value. Alternatively, the first distance S can vary based on the vehicle's speed. In this case, the first distance S increases as the vehicle speed increases. The vehicle speed is determined based on the vehicle state information 220.

[0121] 3-3-4. Step S140

[0122] In step S140, the processor 110 searches for a "first minimum point PM1" that is the minimum point of the first risk potential field U1. Specifically, the processor 110 searches for the first minimum point PM1 in a range near the forward gaze point PA.

[0123] More specifically, the processor 110 Figure 14The first search range AS1 is set as shown. The first search range AS1 extends from the forward gaze point PA in the LY direction (lane width direction). The first search range AS1 is set to cover at least the LY direction of the lane LA. The processor 110 then searches for the first minimum point PM1 within the first search range AS1.

[0124] For example, processor 110 sets multiple check points PC1 within first search range AS1. Processor 110 calculates the risk value R at each check point PC1 by referring to the first risk potential field U1. The risk value R at each check point PC1 can be calculated by substituting the position of each check point PC1 into the potential function (fe, fo) that constitutes first risk potential field U1. Processor 110 then determines the check point PC1 with the smallest risk value R as the first minimum point PM1.

[0125] In this way, the first minimum point PM1 is found in the first search range AS1 near the forward gaze point PA. There is no need to calculate the risk value R for the entire lane LA to search for the first minimum point PM1. Therefore, the computational load required to search for the first minimum point PM1 is significantly reduced.

[0126] 3-3-5. Step S150

[0127] In step S150 , the processor 110 calculates a first deviation D1 , which is a deviation in the LY direction between the forward gaze point PA and the first minimum point PM1 .

[0128] 3-3-6. Step S160

[0129] In step S160, the processor 110 performs steering control in a manner that reduces the first deviation D1. Specifically, the processor 110 calculates the target steering angle θt required to reduce the first deviation D1. Typically, the larger the first deviation D1, the larger the target steering angle θt. A function (e.g., a mapping diagram) representing the correspondence between the first deviation D1 and the target steering angle θt is generated in advance. The processor 110 calculates the target steering angle θt corresponding to the first deviation D1 by referring to the function. Then, the processor 110 performs steering control according to the target steering angle θt. The actual steering angle of the vehicle 1 is obtained based on the vehicle state information 220. The processor 110 controls the steering device 31 to steer the wheels to achieve the target steering angle θt.

[0130] In this way, steering control is performed to bring vehicle 1 closer to the first minimum point PM1. The first valley V1 of the first risk potential field U1 corresponds to a temporally continuous set of first minimum points PM1. By performing steering control to bring vehicle 1 closer to the first minimum point PM1, vehicle 1 can follow the first valley V1 of the first risk potential field U1. In other words, risk avoidance control is achieved.

[0131] Reference Figure 15 The reason why lane departure is prevented in this embodiment will be explained. As described above, the forward gaze point PA is set at a position in the direction of travel (X direction) of vehicle 1. At time ta, the first minimum point PM1 in front of vehicle 1 shifts in a direction away from the target object 5. At this time ta, the forward gaze point PA is located to the left of the first valley V1. The turning direction that reduces the first deviation D1 is right. Therefore, vehicle 1 turns right. As vehicle 1 turns right, the forward gaze point PA also turns right.

[0132] At time tb after vehicle 1 turns right, the forward gaze point PA is located to the right of the first valley V1. The steering direction that reduces first deviation D1 is leftward. Consequently, a restoring steering force is generated to restore the vehicle 1's travel direction. As a result, vehicle 1 does not deviate from lane LA but instead returns to a state of traveling parallel to lane LA. Thus, by setting the forward gaze point PA in the vehicle 1's travel direction (X direction), the vehicle 1 is prevented from deviating from lane LA.

[0133] 3-4. Processing When There Are Multiple First Minimum Point Candidates

[0134] As described above, the first minimum point PM1 of the first risk potential field U1 for steering control is searched for in the first search range AS1. At this time, depending on the number of significant digits of the risk value R, it is possible to find multiple candidates for the first minimum point PM1 in the first search range AS1. Hereinafter, the candidates for the first minimum point PM1 in the first search range AS1 are referred to as "first minimum point candidates CM1". For example, in Figure 16 In the example shown, as a result of the superposition of the longitudinal potential field Ux and the obstacle potential field Uo[i], there are two first minimum point candidates CM1 in the first search range AS1.

[0135] If an inappropriate first minimum point candidate CM1 is selected from the plurality of first minimum point candidates CM1 as the first minimum point PM1, inappropriate steering control will be performed. The occupants of the vehicle 1 (typically the driver) will feel uncomfortable with inappropriate risk avoidance control. To achieve risk avoidance control with minimal discomfort, it is important to select an appropriate first minimum point PM1 from the plurality of first minimum point candidates CM1.

[0136] Figure 17This is a conceptual diagram for explaining a method of selecting an appropriate first minimum point PM1 from a plurality of first minimum point candidates CM1. Figure 17 An example of a plurality of first minimum point candidates CM1 in the search process of this cycle (time t) is shown. In this example, there are three first minimum point candidates CM1_a, CM1_b, and CM1_c in the first search range AS1.

[0137] also, Figure 17 This figure shows an example of the history of the first minimum point PM1 determined in the past (times t-3, t-2, and t-1) search processes. In particular, the first minimum point PM1[t-1] in the previous cycle is referred to as the "previous minimum point." The nearest front object 5N is the object closest to the vehicle 1 among the objects 5 located in front of the vehicle 1. In other words, the nearest front object 5N is the object 5 that is the closest target of risk avoidance control. The first steering direction DS is the steering direction away from the nearest front object 5N.

[0138] First, the first minimum point candidate CM1_a lies in a direction close to the nearest front object marker 5N when viewed from the previous minimum point PM1[t-1]. If the first minimum point candidate CM1_a is selected as the current first minimum point PM1[t], vehicle 1 will be steered so as to approach the nearest front object marker 5N. This steering control is inappropriate and may cause discomfort and unease among the occupants of vehicle 1.

[0139] Both first minimum point candidates CM1_b and CM1_c lie in the first steering direction DS, away from the nearest front object 5N, when viewed from the previous minimum point PM1[t-1]. First minimum point candidate CM1_b is close to the previous minimum point PM1[t-1], while first minimum point candidate CM1_c is farther away from the previous minimum point PM1[t-1]. If first minimum point candidate CM1_c were selected as the current first minimum point PM1[t], excessive steering control would occur. Such excessive steering control could potentially cause discomfort to passengers in vehicle 1.

[0140] Therefore, the processor 110 selects the first minimum point candidate CM1_b as the current first minimum point PM1[t]. Generally speaking, the processor 110 selects the first minimum point candidate CM1 that exists in the first steering direction DS away from the nearest front object 5N when viewed from the previous minimum point PM1[t-1] and is closest to the previous minimum point PM1[t-1] as the current first minimum point PM1[t]. This suppresses inappropriate steering control and excessive steering control. In other words, it is possible to suppress discomfort with risk avoidance control (steering control) based on the first risk potential field U1.

[0141] 3-5. Effect

[0142] As described above, according to this embodiment, the first risk potential field U1 is applied to the steering control of the risk avoidance control. Specifically, the steering control is performed so as to follow the first valley V1 of the first risk potential field U1.

[0143] The first valley V1 is a collection of temporally consecutive first minimum points PM1. The first minimum point PM1 is found within the first search range AS1 near the forward gaze point PA. This eliminates the need to calculate the risk value R for the entire lane LA to search for the first minimum point PM1. This significantly reduces the computational load required to search for the first minimum point PM1.

[0144] If there are multiple first minimum point candidates CM1 within the first search range AS1, the first minimum point candidate CM1 that is located in the first steering direction DS away from the nearest front object 5N when viewed from the previous minimum point PM1[t-1] and closest to the previous minimum point PM1[t-1] is selected as the current first minimum point PM1[t]. This prevents inappropriate and excessive steering control. In other words, it is possible to suppress discomfort with risk avoidance control (steering control) based on the first risk potential field U1.

[0145] The first risk potential field U1 includes a longitudinal potential field Ux for guiding vehicle 1 along lane LA. The longitudinal potential field Ux is preferably the vehicle center potential field Ue. The position of the valley Ve of the vehicle center potential field Ue is not fixed but changes dynamically in conjunction with the position of vehicle 1. Because this valley Ve is reflected at the first minimum point PM1, unnecessary or excessive steering control is suppressed. Suppressing unnecessary or excessive steering control means achieving appropriate vehicle behavior to avoid risks. Consequently, the discomfort experienced by vehicle 1 occupants is minimized.

[0146] Furthermore, according to this embodiment, the target steering angle θt (trajectory TR1) of vehicle 1 is uniquely determined based on the first risk potential field U1. As a comparative example, consider a method that generates multiple target trajectories and selects the optimal one. In this comparative example, an evaluation function is required to evaluate each target trajectory, increasing the computational load. In particular, when there are multiple target objects 5, the evaluation function becomes complex, significantly increasing the computational load. On the other hand, according to this embodiment, since such an evaluation function is not required, the computational load is reduced. The reduction in computational load becomes more significant as the number of target objects 5 increases.

[0147] 4. Deceleration control based on risk potential field

[0148] 4-1. Overview of deceleration control

[0149] Figure 18This is a conceptual diagram for explaining deceleration control based on the risk potential field U. Figure 18 Two objects 5[1] and 5[2] are shown in front of vehicle 1. These two objects 5[1] and 5[2] are relatively close to each other. In this situation, even if the above-mentioned steering control is activated, vehicle 1 will pass relatively close to objects 5[1] and 5[2]. As a result, the risk of collision with object 5[1] is not sufficiently reduced, and the passengers of vehicle 1 may feel uneasy.

[0150] Therefore, in Figure 18 In the situations shown in the example, deceleration control can be considered instead of steering control or in combination with steering control. The concept of "suppression amount" is introduced as a basis for determining in what situations deceleration control should be performed and at what deceleration rate.

[0151] Figure 19 This is a conceptual diagram for explaining the suppression amount used in the deceleration control.

[0152] First, the individual gap Gs[i] associated with the object 5[i] is explained. The individual gap Gs[i] is the lateral distance between the vehicle 1 and the object 5[i], and is the lateral distance at which the passengers will not feel uneasy when the vehicle 1 passes by the side of the object 5[i]. That is, the individual gap Gs[i] is the target lateral distance. The individual gap Gs[i] is pre-set for each object 5[i]. The individual gap Gs[i] may also be a specified value that varies depending on the type of object 5. For example, the individual gap Gs (e.g., 3m) when the object 5 is a pedestrian is greater than the individual gap Gs (e.g., 2m) when the object 5 is a parked vehicle. The individual gap Gs[i] may be based on the distribution parameter σy (refer to Figure 5 The information of the cell gap Gs[i] is included in the potential function information 300 described above, for example.

[0153] Next, the corrected gap Gm[i] associated with the object 5[i] will be described. The corrected gap Gm[i] is the lateral distance between the object 5[i] and the valley of the risk potential field U. The corrected gap Gm[i] can be calculated based on the position of the object 5[i] and the risk potential field U.

[0154] The suppression amount ΔG[i] associated with the object marker 5[i] is the difference between the cell gap Gs[i] and the correction gap Gm[i]. That is, the suppression amount ΔG[i] is expressed by the equation: ΔG[i]=Gs[i]-Gm[i].

[0155] When the correction gap Gm[i] is smaller than the individual gap Gs[i], this means that there are other object marks 5[j] near the object mark 5[i], and the individual gap Gs[i] cannot be ensured. In other words, the situation where the correction gap Gm[i] is smaller than the individual gap Gs[i] is equivalent to Figure 18 In such a situation, deceleration control is preferably performed to reduce the risk of collision and the occupant's sense of unease. Therefore, it can be said that the suppression amount ΔG[i] indicates the necessity of deceleration control.

[0156] According to this embodiment, whether to perform deceleration control is determined based on the suppression amount ΔG[i]. Specifically, deceleration control is performed when the suppression amount ΔG[i] is greater than the threshold value Gth. The target deceleration rate At during deceleration control can also be set based on the suppression amount ΔG[i]. For example, the target deceleration rate At (absolute value) can be set to a higher value as the suppression amount ΔG[i] increases.

[0157] In this way, the suppression amount ΔG[i] is used as a criterion for determination in the deceleration control based on the risk potential field U. In order to appropriately perform the deceleration control, it is necessary to appropriately calculate the suppression amount ΔG[i].

[0158] 4-2. Deceleration Control Based on the Second Risk Potential Field

[0159] Figure 20 This is a block diagram for explaining the overview of deceleration control in this embodiment. The second risk potential field U2 is the risk potential field used for deceleration control. The second risk potential field U2 includes at least the sum of the obstacle potential fields Uo[i] set for each object 5[i]. For example, the second risk potential field U2 is represented by the following equation (2).

[0160] [Formula 2]

[0161]

[0162] Processor 110 calculates the corrected gap Gm[i] and the suppression amount ΔG[i] based on the second risk potential field U2. Specifically, the second valley V2 is the valley of risk value R indicated by the second risk potential field U2. The corrected gap Gm[i] is the lateral distance between the object 5[i] and the second valley V2. The suppression amount ΔG[i] is the difference between the single gap Gs[i] and the corrected gap Gm[i].

[0163] Figure 21 An example of the deceleration control of this embodiment is shown. The positional relationship between the two object marks 5[1] and 5[2] is the same as that of the above-mentioned Figure 18The situation is the same. For the sake of simplicity, it is assumed that the obstacle potential field Uo[1] associated with the object 5[1] and the obstacle potential field Uo[2] associated with the object 5[2] have the same size. In addition, it is assumed that the single-unit gap Gs[1] associated with the object 5[1] and the single-unit gap Gs[2] associated with the object 5[2] are the same. Since the second risk potential field U2 only includes the obstacle potential field Uo[i], the position of the second valley V2 of the second risk potential field U2 coincides with the midpoint of the two object marks 5[1] and 5[2]. Therefore, the suppression amounts ΔG[1] and ΔG[2] both become appropriate values ​​that reflect the proximity of the object marks 5[1] and 5[2]. In other words, the overestimation or underestimation of the suppression amounts ΔG[1] and ΔG[2] is suppressed. As a result, unnecessary deceleration control or excessive deceleration control is suppressed.

[0164] As a modified example, the second risk potential field U2 can be identical to the first risk potential field U1 represented by equation (1). However, in this modified example, the suppression amount ΔG may be excessive. If the suppression amount ΔG is excessive, deceleration control may be unnecessarily initiated, or the target deceleration rate Δt during deceleration control may be excessive. Such unnecessary or excessive deceleration control may cause discomfort to the occupants of vehicle 1 (typically the driver). In this sense, the second risk potential field U2 preferably includes only the obstacle potential field Uo[i].

[0165] It should be noted that the second risk potential field U2 is only used for the calculation of the suppression amount ΔG[i] and is not used for steering control.

[0166] 4-3. Processing Flow

[0167] Figure 22 This is a flowchart showing the processing related to the deceleration control in this embodiment. Figure 22 The processing flow shown is repeatedly executed at a fixed cycle.

[0168] 4-3-1. Step S210

[0169] In step S210, the processor 110 performs the above-mentioned information acquisition process. That is, the processor 110 acquires the driving environment information 200 based on the detection results of the sensor group 20. The driving environment information 200 is stored in the storage device 120. It should be noted that step S210 can be combined with Figure 11 The same as step S110 in .

[0170] 4-3-2. Step S220

[0171] In step S220, processor 110 sets a second risk potential field U2. Second risk potential field U2 comprises the sum of obstacle potential fields Uo[i] (see equation (2)). Based on driving environment information 200 and potential function information 300, processor 110 sets obstacle potential field Uo[i] for each object 5[i]. Processor 110 then sets second risk potential field U2 by superimposing the obstacle potential fields Uo[i] set for object 5[i].

[0172] Figure 23 This is a flowchart showing an example of the processing in step S220.

[0173] In step S221, the processor 110 determines whether there is an object 5 in front of the vehicle 1 based on the object information 250. In other words, the processor 110 determines whether the object 5 is recognized in the area in front of the vehicle 1. If the object 5 in front of the vehicle 1 is recognized (step S221; yes), the process proceeds to step S222. Otherwise (step S221; no), step S220 ends. It should be noted that step S221 can be combined with Figure 12 The same as step S122 in .

[0174] In step S222, the processor 110 determines whether the margin time T to the identified object 5 is less than the first time threshold Tth1. If the margin time T is less than the first time threshold Tth1 (step S222; yes), the process proceeds to step S223. Otherwise (step S222; no), step S220 ends. It should be noted that step S222 can be combined with Figure 12 The same as step S123 in .

[0175] In step S223, processor 110 sets an obstacle potential field Uo associated with the identified object 5 based on driving environment information 200 and potential function information 300. Processor 110 then adds this obstacle potential field Uo to the second risk potential field U2. Thus, when vehicle 1 approaches object 5 to a certain extent, obstacle potential field Uo associated with this object 5 is added to the second risk potential field U2.

[0176] 4-3-3. Step S230

[0177] In step S230, processor 110 determines whether the margin time T is less than a second time threshold Tth2. The second time threshold Tth2 (e.g., approximately 4 to 5 seconds) is less than the first time threshold Tth1 described above. If the margin time T is less than the second time threshold Tth2 (step S230: Yes), processing proceeds to step S240. Otherwise (step S230: No), processing for this cycle ends. It should be noted that "the margin time T is less than the second time threshold Tth2" is the first operating condition for deceleration control.

[0178] 4-3-4. Step S240

[0179] In step S240, the processor 110 searches for the minimum point PM2 of the second risk potential field U2. Specifically, the processor 110 searches for the minimum point PM2 in a range near the object 5[i].

[0180] Reference Figure 24 , the search for the minimum point PM2 is described. The processor 110 Figure 24 The second search range AS2 is set as shown. The second search range AS2 is the range between the position of object 5[i] and the position of the cell gap Gs[i] from object 5[i]. The position of object 5[i] is obtained based on object information 250. The cell gap Gs[i] is obtained based on potential function information 300. The processor 110 sets the second search range AS2 based on the driving environment information 200 and the potential function information 300.

[0181] Furthermore, processor 110 sets multiple check points PC2 within second search range AS2. Processor 110 calculates a risk value R at each check point PC2, referring to the second risk potential field U2. The risk value R at each check point PC2 can be calculated by substituting the position of each check point PC2 into the obstacle potential function fo that constitutes second risk potential field U2. The minimum point PM2 is the check point PC2 at which the risk value R is minimized.

[0182] In this way, the minimum point PM2 is found in the second search range AS2 near the object 5[i]. There is no need to calculate the risk value R for the entire lane LA to search for the minimum point PM2. Therefore, the computational load required to search for the minimum point PM2 is greatly reduced.

[0183] 4-3-5. Step S250

[0184] In step S250, processor 110 determines whether a minimum point PM2 (i.e., a second valley V2) exists within second search range AS2. If no minimum point PM2 exists within second search range AS2 (step S250 returns No), this indicates that there is a sufficient distance between object 5[i] and other object 5[j]. In this case, processor 110 determines that deceleration control is not necessary and terminates processing in this cycle.

[0185] On the other hand, if the minimum point PM2 exists in the second search range AS2 (step S250; Yes), this means that another object 5[j] is located near the object 5[i], and the cell gap Gs[i] cannot be ensured. In this case, the process proceeds to step S260. It should be noted that the presence of the minimum point PM2 in the second search range AS2 is the second operating condition for the deceleration control.

[0186] 4-3-6. Step S260

[0187] In step S260, processor 110 calculates the suppression amount ΔG[i] associated with object marker 5[i]. Specifically, processor 110 calculates the lateral distance between object marker 5[i] and minimum point PM2 as the correction gap Gm[i]. Processor 110 then calculates the difference between the individual gap Gs[i] and the correction gap Gm[i] as the suppression amount ΔG[i].

[0188] 4-3-7. Step S270

[0189] In step S270, processor 110 determines whether the suppression amount ΔG[i] is greater than threshold Gth. If the suppression amount ΔG[i] is greater than threshold Gth (step S270: Yes), processing proceeds to step S280. Otherwise (step S270: No), processing in this cycle ends. "Suppression amount ΔG[i] greater than threshold Gth" is the third operating condition for deceleration control.

[0190] 4-3-8. Step S280

[0191] In step S280, processor 110 performs deceleration control. For example, processor 110 sets a target deceleration At based on the suppression amount ΔG[i]. In this case, the target deceleration At (absolute value) is set to increase as the suppression amount ΔG[i] increases. A function (e.g., a map) is pre-generated that represents the correspondence between the suppression amount ΔG and the target deceleration At. Processor 110 calculates the target deceleration At corresponding to the suppression amount ΔG[i] by referring to this function.

[0192] Then, the processor 110 performs deceleration control according to the target deceleration At. The speed of the vehicle 1 is obtained based on the vehicle state information 220. The processor 110 controls the braking device 33 to achieve the target deceleration At.

[0193] 4-4. Processing when there are multiple minimum points

[0194] As described above, the second valley V2 (minimum point PM2) of the second risk potential field U2 used for deceleration control is searched for within the second search range AS2. At this point, it is possible to find multiple minimum points PM2 within the second search range AS2. For convenience, the multiple minimum points PM2 within the second search range AS2 are referred to as "minimum point candidates CM2."

[0195] If an inappropriate minimum point candidate CM2 is selected from the plurality of minimum point candidates CM2 as the minimum point PM2, the suppression amount ΔG[i] also becomes inappropriate, resulting in inappropriate deceleration control. The occupants of vehicle 1 (typically the driver) experience discomfort with inappropriate risk avoidance control. To achieve risk avoidance control with minimal discomfort, it is important to select an appropriate minimum point PM2 from the plurality of minimum point candidates CM2.

[0196] Figure 25 This is a conceptual diagram for explaining a method of selecting an appropriate minimum point PM2 from a plurality of minimum point candidates CM2. Figure 25 In the example shown, two minimum point candidates CM2_a and CM2_b exist in the second search range AS2. Object 5[i] is the target of deceleration control. Minimum point candidate CM2_a is close to object 5[i], while minimum point candidate CM2_b is far from object 5[i].

[0197] If a minimum point candidate CM2_b located far from object marker 5[i] is selected as minimum point PM2 despite the presence of a minimum point candidate CM2_a located close to object marker 5[i], the correction gap Gm[i] becomes excessively large. In other words, the suppression amount ΔG[i] is underestimated. When the suppression amount ΔG[i] is underestimated, deceleration control is inadequate. The inadequate deceleration control can cause discomfort and anxiety among the occupants of vehicle 1.

[0198] Therefore, processor 110 selects candidate minimum point CM2_a as the minimum point PM2. Generally speaking, processor 110 selects the candidate minimum point CM2 closest to object marker 5[i] among multiple candidate minimum points CM2 as the minimum point PM2. As a result, an appropriate suppression amount ΔG[i] is obtained, enabling appropriate deceleration control. In other words, it is possible to suppress discomfort or anxiety associated with risk avoidance control (deceleration control) based on the second risk potential field U2.

[0199] 4-5. Effect

[0200] As described above, according to this embodiment, the second risk potential field U2 is used for deceleration control in risk avoidance control. Specifically, the minimum point PM2 (second valley V2) of the second risk potential field U2 is searched for, and the suppression amount ΔG[i] is calculated based on the positional relationship between this minimum point PM2 and the object 5[i]. This suppression amount ΔG[i] is then used as a criterion for determining whether to perform deceleration control.

[0201] The minimum point PM2 is found in the second search range AS2 near the target object 5[i]. If the minimum point PM2 does not exist in the second search range AS2, deceleration control is unnecessary, and the calculation of the correction gap Gm[i] and the suppression amount ΔG[i] is not performed. This reduces the computational load.

[0202] If multiple minimum point candidates CM2 exist within the second search range AS2, the one closest to the object mark 5[i] is selected as the minimum point PM2. This results in an appropriate suppression amount ΔG[i], enabling appropriate deceleration control. This reduces discomfort and anxiety associated with risk avoidance control (deceleration control) based on the second risk potential field U2.

[0203] Preferably, the second risk potential field U2 includes only the obstacle potential field Uo[i]. In this case, the position of the minimum point PM2 is determined solely based on the positional relationship of the object marker 5[i]. Since the suppression amount ΔG[i] is calculated based on this minimum point PM2, an appropriate suppression amount ΔG[i] that reflects the proximity of the object marker 5[i] can be obtained. As a result, unnecessary or excessive deceleration control is suppressed, thereby minimizing discomfort experienced by the occupants of the vehicle 1.

[0204] 5. Combination of steering control and deceleration control

[0205] A combination of steering control and deceleration control is also possible. The first risk potential field U1 is applied to steering control, and the second risk potential field U2 is applied to deceleration control. This achieves both the effects described in Section 3 and the effects described in Section 4.

Claims

1. A driving assistance system for assisting driving of a vehicle, the driving assistance system comprising: a storage device storing driving environment information representing a driving environment of the vehicle; as well as a processor that executes risk avoidance control for reducing a risk of collision with an object in front of the vehicle based on the driving environment information, The risk potential field shows the risk value as a function of position, The obstacle potential field is a risk potential field in which the risk value is the largest at the location of the object and decreases as the distance from the object increases. The longitudinal potential field is the risk potential field in which the valley of the risk value extends in the direction of the lane length. The risk avoidance control includes a deceleration control for decelerating the vehicle, The risk potential field used for the deceleration control includes at least the obstacle potential field, The search range is the range between the position of the object and a position with a predetermined gap from the object. The processor is configured to: Based on the driving environment information, the risk potential field for the deceleration control is set by superimposing only the obstacle potential field set for each of the object markers without using the longitudinal potential field; searching for a minimum point of the risk potential field for the deceleration control from the search range; The deceleration control is performed based on the positional relationship between the minimum point and the target object. When there are multiple minimum points in the search range, the processor selects a minimum point closest to the target object among the multiple minimum points as the minimum point.

2. The driving assistance system according to claim 1, wherein: The processor performs the deceleration control when a corrected gap, which is a lateral distance between the object and the minimum point, is smaller than the prescribed gap and a difference between the prescribed gap and the corrected gap is larger than a threshold.

3. The driving assistance system according to claim 2, wherein: The processor sets a target deceleration such that the larger the difference between the predetermined gap and the corrected gap, the higher the target deceleration, and performs the deceleration control according to the target deceleration.

Images