Driving assistance system, driving assistance method, and computer-readable recording medium

Abstract

The present invention relates to a driving assistance system, a driving assistance method, and a computer-readable recording medium. The present invention provides a driving assistance system that can suppress the inconvenience and anxiety caused to the driver and reduce the collision risk caused by an object located in front of a vehicle. According to the driving assistance system, risk object information related to a risk object that exists as a risk for the vehicle is extracted from information about the surrounding conditions of the vehicle. In addition, influencing factor information related to an influencing factor is obtained, which is a factor that exists independently of the risk object and affects the collision risk. Then, a risk value that quantifies the collision risk is determined based on the risk object information and the influencing factor information, and based on the determined risk value, the operation amount of the actuator is determined in a manner that reduces the collision risk.

Description

Technical Field

[0001] The present invention relates to a driving assistance system for assisting driving of a vehicle, a driving assistance method, and a computer-readable recording medium storing a driving assistance program. Background Art

[0002] If a pedestrian or cyclist suddenly emerges from the blind spot of a wall or parked vehicle, conventional AEBS brakes may not be able to adequately decelerate and avoid the situation, potentially leading to a collision. Therefore, Patent Document 1 proposes a "risk field method" that identifies the blind spot and defines the predicted collision speed after AEBS activation based on the distance, lateral clearance, and relative speed between the vehicle and the blind spot, assuming a hypothetical pedestrian emerges. By decelerating or performing lateral avoidance until this assumed potential risk value (predicted collision speed) reaches zero, the vehicle can safely navigate the blind spot.

[0003] Prior art literature

[0004] Patent Literature

[0005] Patent Document 1: Japanese Patent Application Laid-Open No. 2017-206117

[0006] For example, when steering to avoid a potential hazard on a typical road, the road is narrow, making adequate avoidance impossible. In this case, the driver must rely on deceleration to avoid the hazard. However, aiming for zero risk results in significant deceleration, which can be inconvenient for the driver. Conversely, constantly executing the same avoidance control can cause driver anxiety depending on the situation. Summary of the Invention

[0007] The present invention has been completed in view of the above-mentioned problems, and its purpose is to provide a driving assistance system, a driving assistance method and a computer-readable recording medium storing a driving assistance program that can suppress the trouble and anxiety caused to the driver and reduce the risk of collision caused by objects existing in front of the vehicle.

[0008] The driving assistance system of the present invention comprises: one or more memories storing one or more programs; and one or more processors coupled to the one or more memories. When the one or more programs are executed, the one or more processors perform the following first to fourth processes. The first process is a process of extracting risk object information related to risk objects that exist as a risk that causes the vehicle to have a collision risk from information about the surrounding conditions of the vehicle. The second process is a process of obtaining influencing factor information related to influencing factors, which are factors that exist independently of the risk objects and are factors that affect the collision risk. The third process is a process of determining a risk value that quantifies the collision risk based on the risk object information and the influencing factor information. The fourth process is a process of determining the operation amount of the actuator that controls the movement of the vehicle in a manner that reduces the collision risk based on the determined risk value.

[0009] According to the driving assistance system configured as described above, a risk value that quantifies the collision risk is determined based on risk object information and influencing factor information. Based on this risk value, the actuator operation amount is determined to reduce the collision risk. While risk object information is information related to risk objects that pose a risk of collision to the vehicle, influencing factor information is information related to influencing factors that exist independently of the risk objects and affect the collision risk. By incorporating influencing factor information into risk object information to determine the risk value, it is possible to appropriately intervene in actuator operation to reduce the collision risk.

[0010] As a first embodiment of the driving assistance system of the present invention, one or more processors, when executing one or more programs, may extract information related to a potential risk object located in front of the vehicle and forming a blind spot when viewed from the vehicle as risk object information. In this first embodiment, information related to the surrounding environment of the potential risk object may also be obtained as influencing factor information. Alternatively, in the first embodiment, information related to a moving object located behind the potential risk object may also be obtained as influencing factor information. Alternatively, in the first embodiment, information related to dynamic factors acting on the blind spot formed by the potential risk object may also be obtained as influencing factor information. Alternatively, in the first embodiment, information related to the time and location of the detection of the potential risk object may also be obtained as influencing factor information.

[0011] As a second aspect of the driving assistance system of the present invention, one or more processors may extract, as risk object information, information related to a visible risk object located in front of the vehicle and potentially colliding with the vehicle. In this second aspect, if the visible risk object is a parked vehicle, information related to the presence or absence of a driver in the parked vehicle may also be obtained as influencing factor information. Alternatively, in this second aspect, information related to the state of the road where the visible risk object was detected may also be obtained as influencing factor information. Alternatively, in this second aspect, information related to the time and location of the detection of the visible risk object may also be obtained as influencing factor information.

[0012] The driving assistance method of the present invention includes the following first to fourth steps. The first step is a step of extracting risk marker information related to risk markers that exist and pose a risk of collision to the vehicle from information about the surrounding conditions of the vehicle. The second step is a step of obtaining influencing factor information related to influencing factors, which are factors that exist independently of the risk markers and affect the collision risk. The third step is a step of determining a risk value that quantifies the collision risk based on the risk marker information extracted in the first step and the influencing factor information obtained in the second step. Furthermore, the fourth step is a step of determining the amount of operation of an actuator that controls the movement of the vehicle in a manner that reduces the collision risk based on the risk value determined in the third step.

[0013] The driving assistance program stored in the computer-readable recording medium of the present invention causes the computer to execute the following first to fourth processes. The first process is a process of extracting risk marker information related to risk markers that exist as a risk marker that causes the vehicle to have a collision risk from information about the surrounding conditions of the vehicle. The second process is a process of obtaining influencing factor information related to influencing factors, which are factors that exist independently of the risk markers and are factors that affect the collision risk. The third process is a process of determining a risk value that quantifies the collision risk based on the risk marker information extracted in the first process and the influencing factor information obtained in the second process. And, the fourth process is a process of determining the operation amount of the actuator that controls the movement of the vehicle in a manner that reduces the collision risk based on the risk value determined in the third process.

[0014] Effects of the Invention

[0015] The driving assistance system, driving assistance method, and computer-readable recording medium storing a driving assistance program according to the present invention determine a risk value by incorporating influencing factor information into risky object information. This allows for appropriate intervention in actuator operations to reduce collision risk. This reduces the driver's annoyance and anxiety, while also reducing the risk of collision with objects ahead of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a conceptual diagram for explaining the outline of potential risk avoidance control among driving assistance controls performed by the driving assistance system according to the embodiment of the present invention.

[0017] Figure 2 This is a conceptual diagram for explaining the outline of potential risk avoidance control among driving assistance controls performed by the driving assistance system according to the embodiment of the present invention.

[0018] Figure 3 This is a conceptual diagram for explaining the outline of explicit risk avoidance control in driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0019] Figure 4 This is a conceptual diagram for explaining the outline of explicit risk avoidance control in driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0020] Figure 5 This is a block diagram showing a configuration example of a driving assistance system according to an embodiment of the present invention and a vehicle to which the driving assistance system is applied.

[0021] Figure 6 This is a block diagram showing processing executed by a processor according to an embodiment of the present invention.

[0022] Figure 7 This is a conceptual diagram for explaining a first example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0023] Figure 8 This is a conceptual diagram for explaining a first example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0024] Figure 9 This is a conceptual diagram for explaining a second example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0025] Figure 10 This is a conceptual diagram for explaining a second example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0026] Figure 11 This is a conceptual diagram for explaining a third example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0027] Figure 12This is a conceptual diagram for explaining a third example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0028] Figure 13 This is a conceptual diagram for explaining a fourth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0029] Figure 14 This is a conceptual diagram for explaining a fourth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0030] Figure 15 This is a conceptual diagram for explaining a fifth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0031] Figure 16 This is a conceptual diagram for explaining a fifth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0032] Figure 17 This is a conceptual diagram for explaining a fifth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0033] Figure 18 This is a conceptual diagram for explaining a sixth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0034] Figure 19 This is a conceptual diagram for explaining a sixth example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0035] Figure 20 This is a conceptual diagram for explaining a seventh example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0036] Figure 21 This is a conceptual diagram for explaining a seventh example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0037] Figure 22 This is a conceptual diagram for explaining a seventh example of driving assistance control performed by the driving assistance system according to the embodiment of the present invention.

[0038] Description of reference numerals:

[0039] 10 sensor groups;

[0040] 11 autonomous sensors;

[0041] 12 vehicle status sensors;

[0042] 13. Position sensor;

[0043] 20 control devices;

[0044] 21 processors;

[0045] 22 memory;

[0046] 23 driver assistance programs;

[0047] 24 driving environment information;

[0048] 25 risk information;

[0049] 30 vehicle actuators;

[0050] 31 steering actuator;

[0051] 32 driving actuator;

[0052] 33 brake actuator;

[0053] 100 driver assistance systems;

[0054] VH vehicles;

[0055] CL1, CL2 dividing line;

[0056] VP imaginary pedestrian;

[0057] There are actual pedestrians in the RP;

[0058] BW retaining wall;

[0059] DV driver;

[0060] BL brake light;

[0061] RC road cones;

[0062] PV, PVL, PVS parking vehicles;

[0063] PR, PR11, PR12, PR21, PR22, PR31, PR32, PR41, PR42 potential risk items;

[0064] ER, ER51, ER52, ER53, ER61, ER62, ER71, ER72, ER73 are risk indicators;

[0065] Influencing factors of IF01, IF02, IF03, IF04, IF11, IF22, IF32, IF411, IF412, IF421, IF422, IF52, IF531, IF532, IF62, IF721, IF722, IF731, IF732, IF733;

[0066] RF01, RF02, RF03, RF04, RF11, RF12, RF21, RF22, RF31, RF32, RF41, RF42, RF51, RF52, RF53, RF61, RF62, RF71, RF72, RF73 risk potential fields;

[0067] TR01, TR02, TR03, TR04, TR11, TR12, TR21, TR22, TR31, TR32, TR41, TR42, TR51, TR52, TR53, TR61, TR62, TR71, TR72, TR73 target trajectories. DETAILED DESCRIPTION

[0068] The following describes embodiments of the present invention with reference to the accompanying drawings. In the embodiments described below, when reference is made to numbers, quantities, amounts, ranges, and other values for various elements, the present invention is not limited to such values unless otherwise specified or clearly established in principle. Furthermore, with respect to structures and the like described in the embodiments described below, unless otherwise specified or clearly established in principle, such structures and the like are not necessarily required for the present invention.

[0069] 1. Overview of the Driving Assistance System of the Present Embodiment

[0070] 1-1. Overview of Driving Assistance Control

[0071] The driving assistance system of this embodiment performs driving assistance control to assist the vehicle's driving by avoiding the risk of collision with objects in front of it. The collision risks that the vehicle should avoid are classified as potential and manifest. Potential risks are potential collision risks in the vehicle's blind spot. Manifest risks are obvious collision risks, such as the possibility of running over onto pedestrians on the road. The driving assistance system of this embodiment avoids both types of collision risks.

[0072] In driver assistance control, a risk value that quantifies the risk of collision is used. The risk value is given as a distribution in the vehicle coordinate system or the absolute coordinate system. The distribution of this risk value is defined as a risk potential field. Typically, the risk value is defined based on information related to the object to be avoided, such as the position of the object to be avoided, the distance to the object, the type of the object, the size of the object, and the displacement speed of the object. It should be noted that coordinate transformations can be performed between the vehicle coordinate system and the absolute coordinate system.

[0073] When the collision risk is a latent risk, the target object is a hypothetical object lurking in the blind spot of an object generating a potential risk (hereinafter referred to as a potential risk object). In this case, hypothetical information associated with the potential risk object is provided as information related to the target object for defining the risk value. Therefore, when the collision risk is a latent risk, the distribution of risk values is associated with the potential risk object. On the other hand, when the collision risk is an explicit risk, the target object is the object generating the explicit risk (hereinafter referred to as the explicit risk object). In this case, the risk value is determined based on information related to the explicit risk object, and the distribution of risk values is associated with the explicit risk object.

[0074] As described above, the risk value for driving assistance control is determined based on information related to potential risk objects or manifest risk objects. This information is hereinafter referred to as risk object information. Risk object information is information related to risk objects that pose a risk of collision to the vehicle and is extracted from surrounding condition information acquired by autonomous sensors mounted on the vehicle. However, in the driving assistance system of this embodiment, the information used to determine the risk value is not limited to risk object information.

[0075] The driving assistance system of this embodiment uses information related to factors that exist independently of risk objects and influence collision risk to determine the risk value. Collision risk is determined not only by the risk object but also by various factors surrounding it. Hereinafter, factors influencing collision risk are referred to as influencing factors, and information related to these influencing factors is referred to as influencing factor information. It can also be said that risk object information is the basic value that determines the risk value, while influencing factor information provides correction terms or coefficients that correct this basic value.

[0076] Driving assistance control involves controlling the vehicle to avoid collision risks. This risk-avoidance vehicle control includes at least one of braking control, which operates a brake actuator to brake the vehicle, and steering control, which operates a steering actuator to steer the vehicle. The aforementioned risk value, more specifically, a risk potential field representing a distribution of risk values, is used to determine the amount of operation for each actuator.

[0077] Hereinafter, driving assistance control performed with potential risks as the avoidance target will be referred to as potential risk avoidance control, and driving assistance control performed with explicit risks as the avoidance target will be referred to as explicit risk avoidance control. The following sections describe potential risk avoidance control and explicit risk avoidance control in more detail.

[0078] 1-2. Potential risk avoidance control

[0079] Figure 1 and Figure 2 : is a conceptual diagram for explaining the outline of potential risk avoidance control performed by the driving assistance system 100 of this embodiment. Figure 1 and Figure 2 The figure depicts a vehicle VH traveling in a lane defined by two dividing lines CL1 and CL2. A fork in the road connects to the right side of the lane. The existence of the fork can be obtained from map information. A building is identified in front of the fork as a potential risk object PR. A potential risk object PR is defined as an object that exists in front of the vehicle VH and forms an area that is a blind spot when viewed from the vehicle VH. More specifically, an area that is a blind spot when viewed from the vehicle VH refers to an area that is a blind spot for the autonomous sensors mounted on the vehicle VH. The potential risk object PR itself can be identified by the autonomous sensors.

[0080] The potential risk object PR creates a blind spot at the junction that is not visible from the vehicle VH. In the potential risk avoidance control, it is assumed that there is a virtual pedestrian VP behind the potential risk object PR. Then, risk potential fields RF01 and RF02 are generated that extend around the virtual pedestrian VP. For example, Figure 1 and Figure 2 As shown, the risk potential fields RF01 and RF02 can be represented by contour lines connecting sets of points with the same risk value. In this example, the contour lines closer to the center have a larger risk value, and the contour lines further out have a smaller risk value. It should be noted that Figure 1 and Figure 2The risk potential fields RF01 and RF02 shown represent regions with risk values greater than or equal to a certain value. Risk values greater than or equal to a certain value are risk values that the vehicle VH should avoid. This is also common to the other figures used in this application, where regions with risk values greater than or equal to a certain value are depicted as risk potential fields using contour lines.

[0081] The risk value for each location forming the risk potential fields RF01 and RF02 is determined by the driving assistance system 100. During potential risk avoidance control, the driving assistance system 100 extracts risk object information related to potential risk objects PR from the surrounding condition information of the vehicle VH acquired by autonomous sensors, and also obtains influencing factor information related to influencing factors IF01 and IF02. Other examples of potential risk objects PR include block walls and walls at intersections and T-junction corners, and parked vehicles on the roadside. Specific examples of influencing factors will be described in the embodiments of potential risk avoidance control described later.

[0082] The driving assistance system 100 determines the risk value based on the risk object information and the influencing factor information. The distribution of the determined risk value is the risk potential field RF01, RF02. Figure 1 and Figure 2 , the distribution in the vehicle coordinate system is shown with the lateral direction of the vehicle VH as the X-axis and the traveling direction of the vehicle VH as the Y-axis. This is also common to other drawings used in this application.

[0083] In the case of a common potential risk object marker PR, the risk object marker information is the same, so the difference in the magnitude of the risk potential field RF01, RF02 is generated according to the difference in influencing factors IF01, IF02. Figure 2 The risk potential field RF02 shown is Figure 1 The risk potential field RF01 is shown to be large. This is because Figure 2 The influencing factors IF02 and Figure 1 Compared with the influencing factor IF01, it has a greater impact on the collision risk.

[0084] The driving assistance system 100 generates target trajectories TR01 and TR02 for the vehicle VH based on the risk potential fields RF01 and RF02. The target trajectories TR01 and TR02 are the trajectories along which the vehicle VH moves along the target route. They include a set of target positions of the vehicle VH in the vehicle coordinate system and a target speed at each target point. Typically, the target trajectories TR01 and TR02 are generated with the vehicle VH traveling at the legal speed in the center of the lane. Figure 1 In the example shown, the risk potential field RF01 is narrow, so the target trajectory TR01 can be drawn in a way that does not interfere with the risk potential field RF01. Figure 2 In the example shown, the risk potential field RF02 expands in the direction of the lane in which the vehicle VH is traveling. Figure 2 In the example shown, a target trajectory TR02 is generated that bypasses the risk potential field RF02.

[0085] Driving assistance system 100 determines the operating amounts of each actuator so that vehicle VH follows target trajectories TR01 and TR02. Target trajectories TR01 and TR02 are generated based on risk potential fields RF01 and RF02. Therefore, ensuring vehicle VH follows target trajectories TR01 and TR02 means determining the operating amounts of each actuator to reduce the risk of collision with potential risk objects PR.

[0086] This overview of potential risk avoidance control incorporates influencing factor information into risk object information to determine risk potential fields RF01 and RF02, enabling appropriate intervention in actuator operations to reduce collision risk. This minimizes driver inconvenience and anxiety, while also reducing the risk of collision caused by a potential risk object PR located in front of the vehicle VH.

[0087] 1-3. Explicit risk avoidance control

[0088] Figure 3 and Figure 4 This is a conceptual diagram for explaining the outline of the explicit risk avoidance control performed by the driving assistance system 100 of this embodiment. Figure 3 and Figure 4 The vehicle VH is depicted traveling in a lane defined by two dividing lines CL1 and CL2. The roadside strip is located between the left dividing line CL1 and the outer retaining wall BW. Figure 3 and Figure 4 In the embodiment, a pedestrian RP is recognized as a significant risk object ER walking near the dividing line CL1 in the roadside strip. This pedestrian RP is not a hypothetical pedestrian but an actual pedestrian recognized by the autonomous sensor.

[0089] Around the explicit risk marker ER, risk potential fields RF03 and RF04 are generated, extending from the explicit risk marker ER. In the risk potential fields RF03 and RF04 based on the explicit risk marker ER, the contours closer to the center have higher risk values, while the contours farther out have lower risk values.

[0090] The risk value for each location forming the risk potential fields RF03 and RF04 is determined by the driving assistance system 100. During explicit risk avoidance control, the driving assistance system 100 extracts risk object information related to explicit risk objects ER from the surrounding condition information of the vehicle VH acquired by autonomous sensors, and also obtains influencing factor information related to influencing factors IF03 and IF04. Other examples of explicit risk objects ER include bicycles, two-wheeled vehicles, and parked vehicles on the roadside. Other examples of explicit risk objects ER include bicycles, two-wheeled vehicles, and leading vehicles in the lane. Specific examples of influencing factors related to explicit risk objects ER will be described in the embodiments of explicit risk avoidance control described later.

[0091] The driving assistance system 100 determines the risk value based on the risk object information and the influencing factor information. The distribution of the determined risk value is the risk potential field RF03 and RF04. When the risk object ER is common, the risk object information is the same. Therefore, the difference in the influencing factors IF03 and IF04 will generate the difference in the size of the risk potential field RF03 and RF04. For example, Figure 4 The risk potential field RF04 shown is Figure 3 The risk potential field RF03 is shown to be large. This is because Figure 4 The influencing factors IF04 and Figure 1 Compared with the influencing factor IF03, it has a greater impact on the collision risk.

[0092] The driving assistance system 100 generates target trajectories TR03 and TR04 of the vehicle VH based on the risk potential fields RF03 and RF04. Figure 3 In the example shown, the risk potential field RF03 is narrow, so in order to prevent the target trajectory TR03 from interfering with the risk potential field RF03, the target trajectory TR03 can be slightly bulged to the right. Figure 4 In the example shown, the risk potential field RF04 extends significantly to the middle of the lane in which the vehicle VH is traveling. Figure 4 In the example shown, a target trajectory TR04 is generated that largely bypasses the risk potential field RF04 to the right.

[0093] Driving assistance system 100 determines the operating amounts of each actuator so that vehicle VH follows target trajectories TR03 and TR04. Target trajectories TR03 and TR04 are generated based on risk potential fields RF03 and RF04. Therefore, ensuring that vehicle VH follows target trajectories TR03 and TR04 means determining the operating amounts of each actuator so as to reduce the collision risk associated with the displayed risk object ER.

[0094] This schematically illustrated explicit risk avoidance control incorporates influencing factor information into risk object information to determine risk potential fields RF03 and RF04, enabling appropriate intervention in actuator operations to reduce collision risk. This minimizes driver inconvenience and anxiety, while also reducing the collision risk associated with explicit risk objects ER located in front of vehicle VH.

[0095] 2. Configuration and Functions of the Driving Assistance System of This Embodiment

[0096] 2-1. Configuration of the Driving Assistance System

[0097] Figure 5 This figure shows an example configuration of a driving assistance system 100 according to this embodiment and a vehicle VH to which the driving assistance system 100 is applied. The vehicle VH includes a control device 20 that controls the vehicle VH; a sensor group 10 that inputs information to the control device 20; and a vehicle actuator 30 that operates based on signals output from the control device 20. The control device 20, the sensor group 10, and the vehicle actuator 30 are connected via an in-vehicle network. The driving assistance system 100 includes at least the control device 20. However, the driving assistance system 100 may also include the sensor group 10 in addition to the control device 20. Furthermore, the driving assistance system 100 may also include the vehicle actuator 30.

[0098] The sensor group 10 includes an autonomous sensor 11, a vehicle state sensor 12, and a position sensor 13. The autonomous sensor 11 is a sensor that obtains information about the surrounding conditions of the vehicle, including the area in front of the vehicle VH. The autonomous sensor 11 includes at least one of a camera, a millimeter-wave radar, and a LiDAR (Laser Imaging Detection and Ranging). Based on the information obtained by the autonomous sensor 11, processing is performed such as sensing objects around the vehicle VH, measuring the relative position or relative speed of the sensed objects relative to the vehicle VH, and recognizing the shape of the sensed objects. The vehicle state sensor 12 is a sensor that obtains information related to the movement of the vehicle VH. For example, the vehicle state sensor 12 includes at least one of a wheel speed sensor, an acceleration sensor, a yaw rate sensor, and a steering angle sensor. The position sensor 13 is used to obtain information related to the current position of the vehicle VH. As an example of the position sensor 13, a GPS (Global Positioning System) receiver is shown.

[0099] The vehicle actuator 30 controls the movement of the vehicle VH. The vehicle actuator 30 includes a steering actuator 31 for steering the vehicle VH, a drive actuator 32 for driving the vehicle VH, and a brake actuator 33 for braking the vehicle VH. Examples of the steering actuator 31 include power steering systems, steer-by-wire systems, and rear-wheel steering systems. Examples of the drive actuator 32 include engines, EV systems, and hybrid systems. Examples of the brake actuator 33 include hydraulic brakes and regenerative brakes.

[0100] The control device 20 is an ECU (Electronic Control Unit) mounted on the vehicle VH or a collection of multiple ECUs. Alternatively, part of the functions of the control device 20 or all of the functions may be configured on an external server. In this case, the vehicle VH is connected to the server via a mobile communication network. In any case, the control device 20 has one or more processors 21 and one or more memories 22. The memory 22 includes a main storage device and an auxiliary storage device. The memory 22 stores programs that can be executed by the processor 21 and various information associated with the program. The program includes a driving assistance program 23 for enabling the processor 21 to perform the aforementioned driving assistance control. The driving assistance program 23 can be stored in the main storage device or in a computer-readable recording medium serving as an auxiliary storage device. The information stored in the memory 22 includes driving environment information 24 and risk information 25.

[0101] 2-2. Information stored in memory

[0102] The driving environment information 24 is information indicating the driving environment of the vehicle VH. The driving environment information 24 includes, for example, vehicle position information, vehicle status information, and map information. The vehicle position information is information indicating the position and orientation of the vehicle VH obtained from the detection results obtained by the position sensor 13. The vehicle status information is information such as vehicle speed, yaw rate, lateral acceleration, and steering angle obtained from the detection results obtained by the vehicle status sensor 12. Map information includes, for example, lane configuration and road shape. The control device 20 obtains map information of the required area from the map database. The map database can be stored in a specified memory mounted on the vehicle VH or obtained from a server outside the vehicle VH. The driving environment information 24 also includes peripheral condition information indicating the conditions around the vehicle VH.

[0103] The driving environment information 24 also includes surrounding situation information indicating the surrounding situation of the vehicle VH. The surrounding situation information includes information obtained by the autonomous sensor 11, such as image information indicating the surrounding situation of the vehicle VH captured by a camera and measurement information obtained by millimeter-wave radar or LiDAR.

[0104] Surrounding condition information also includes road configuration information. Road configuration information is information related to the relative position of the road configuration surrounding the vehicle VH relative to the vehicle VH. The road configuration surrounding the vehicle VH includes dividing lines and road edge objects. Road edge objects are three-dimensional objects representing the edges of the road, such as curbs, guardrails, walls, and medians. The relative positions of these road configurations can be determined, for example, by analyzing image information captured by a camera.

[0105] The surrounding condition information also includes object information. Object information is information related to objects surrounding the vehicle VH. Object information includes the relative position and relative speed of the object relative to the vehicle VH. For example, the object can be identified by analyzing the image information obtained by the camera, and the relative position of the object can be calculated. In addition, the object can be identified based on radar measurement information, and the relative position and relative speed of the object can be obtained. In addition, the object information includes the size and type of the identified object. The object information can also include the moving direction and moving speed of the object. Moreover, the object information can also include the history of the relative position, relative speed, moving direction and moving speed of the object within a certain period of time in the past. The object includes the aforementioned potential risk objects and obvious risk objects, and the object information includes the aforementioned risk object information.

[0106] Driving environment information 24 also includes influencing factor information. Influencing factor information is information related to factors that influence the collision risk of vehicle VH. There are two types of influencing factor information: information related to factors that influence the collision risk generated by a potential risk object and information related to factors that influence the collision risk generated by the potential risk object. Examples of the influencing factor information include information related to the surrounding environment of the potential risk object, information related to mobile objects behind the potential risk object, information related to dynamic factors affecting the blind spot formed by the potential risk object, and information related to the time and location of the detection of the potential risk object. Examples of the influencing factor information include information related to the presence of a driver in a parked vehicle (if the risk object is a parked vehicle), information related to the state of the road where the risk object was detected, and information related to the time and location of the detection of the risk object. Influencing factor information is stored in association with the risk object information.

[0107] The risk information 25 is information related to the risk potential field on the road on which the vehicle VH is traveling. The distribution of risk values in the vehicle coordinate system or the absolute coordinate system is stored as the risk information 25. The risk value is calculated by the processor 21 based on the risk object information and the influencing factor information.

[0108] 2-3. Processing performed by the processor

[0109] Figure 6 21 is a block diagram showing processes executed by the processor 21 when the driving support program 23 is executed. When the driving support program 23 is executed, the processor 21 executes processes 211 , 212 , 213 , 214 , and 215 .

[0110] First, the processor 21 executes process 211. Through process 211, the surrounding condition information is acquired from the autonomous sensor 11. Strictly speaking, the surrounding condition information detected by the autonomous sensor 11 is temporarily stored in the memory 22, and the temporarily stored surrounding condition information is read out to the processor 21.

[0111] Next, processor 21 executes process 212. Process 212 extracts risk object information related to risk objects presenting a collision risk from the object information included in the surrounding condition information. Whether an object in front of vehicle VH is a risk object is determined based on the object's relative position, relative speed, size, and type. Risk object information is extracted from the surrounding condition information for all risk objects in front of vehicle VH.

[0112] Processor 21 executes process 213 in parallel with process 212. Process 213 acquires influencing factor information associated with the identified risk object from sensor group 10. The influencing factors that affect collision risk vary depending on the type of risk object generating the collision risk. For each identified risk object, processor 21 acquires influencing factor information from sensor group 10 for all influencing factors that affect the collision risk generated by the risk object.

[0113] After executing processes 212 and 213, processor 21 executes process 214. Process 214 determines a risk value that quantifies the collision risk based on the risk object information and the influencing factor information. Processor 21 calculates a basic distribution of risk values based on the risk object information. If the risk object information is identical, the basic distributions of risk values are identical, and the shapes of the risk potential fields represented by contour lines are also identical. Next, processor 21 corrects the distribution of risk values based on the basic distribution based on the influencing factor information. For example, if an influencing factor affects the vehicle in a way that increases the collision risk, processor 21 corrects the distribution of risk values to increase the risk value at each location relative to the basic distribution of risk values. Furthermore, if an influencing factor affects the vehicle in a way that increases the lateral collision risk, processor 21 corrects the distribution of risk values to increase the risk value in the lateral direction relative to the basic distribution of risk values.

[0114] After executing process 214, processor 21 executes process 215. Process 215 determines the actuator operation amount based on the risk value distribution determined in process 214. Specifically, a target trajectory with a low collision risk is generated based on the risk value distribution, and the actuator operation amount is determined to cause vehicle VH to follow the target trajectory. Processor 21 operates vehicle actuator 30 according to the actuator operation amount determined in process 215. The steering of vehicle VH is controlled by operating steering actuator 31 by processor 21. The driving of vehicle VH is controlled by operating drive actuator 32 by processor 21. The braking of vehicle VH is controlled by operating brake actuator 33 by processor 21.

[0115] 3. Examples of Driving Assistance Control

[0116] 3-1. First embodiment

[0117] Figure 7 and Figure 8 This is a conceptual diagram for explaining a first embodiment of the driving assistance control performed by the driving assistance system 100. The first embodiment of the driving assistance control is an embodiment of potential risk avoidance control. In the first embodiment, a branch road is connected to the right side of the driving lane defined by two dividing lines CL1 and CL2. A retaining wall BW is erected on the right side of the driving lane and on both sides of the branch road. Therefore, when viewed from the vehicle VH traveling in the driving lane, a blind spot formed by the retaining wall BW appears at the corner where the branch road connects to the driving lane. The driving assistance system 100 identifies the retaining wall BW that exists in front of the vehicle VH and forms an area that becomes a blind spot when viewed from the vehicle VH as potential risk markers PR11 and PR12.

[0118] In the potential risk avoidance control, it is assumed that a virtual pedestrian VP exists in the blind spot of the potential risk objects PR11 and PR12. The driving assistance system 100 extracts risk object information related to the virtual pedestrian VP from the surrounding situation information and obtains influencing factor information related to the virtual pedestrian VP. The risk object information is object information related to the barrier wall BW as the potential risk objects PR11 and PR12. Figure 7 The examples shown and Figure 8 The examples shown are universal. On the other hand, the influencing factor information is Figure 7 The examples shown and Figure 8 The example shown is different.

[0119] In the first embodiment, the driving assistance system 100 obtains information related to the surrounding environment of the retaining wall BW as the potential risk object PR11, PR12 as the influencing factor information. The surrounding environment of the potential risk object is, for example, a school, a park, a commercial area, a residential area, a factory area, an open space, etc. Figure 7 In the example shown, the influencing factor IF11 that affects the collision risk is the residential area that exists around the potential risk object PR11. Figure 8 In the example shown, factor IF12 that affects collision risk is the presence of a school near potential risk object PR12. The risk of children running out is higher near schools, so factor IF12 has a greater impact on collision risk than factor IF11.

[0120] exist Figure 7 In the example shown, information related to a residential area as the influencing factor IF11 is acquired as the influencing factor information. Figure 7 In the example shown, information related to the school as the influencing factor IF12 is acquired as the influencing factor information. This influencing factor information is information related to the surrounding environment of the potential risk objects PR11 and PR12, and can be acquired based on map information and vehicle position information, for example.

[0121] Based on the risk object information and influencing factor information related to the virtual pedestrian VP, the driving assistance system 100 generates risk potential fields RF11 and RF12, which extend from the virtual pedestrian VP as the center. As described above, the influencing factor IF12, representing a school, has a greater impact on collision risk than the influencing factor IF11, representing a residential area. Furthermore, the collision risk assumed in the first embodiment is the risk caused by the virtual pedestrian VP running off a side road. Therefore, as the collision risk increases, the distribution of risk values spreads further toward the destination of the run-off.

[0122] In the first embodiment, the driving assistance system 100 sets the risk potential fields RF11 and RF12 as ellipses that expand from the branch road to the driving lane. Figure 8 In the example shown, Figure 7 Compared to the risk potential field RF11 in the example shown, the driving assistance system 100 further expands the risk potential field RF12 in the direction from the junction to the driving lane.

[0123] The driving assistance system 100 generates target trajectories TR11 and TR12 of the vehicle VH based on the risk potential fields RF11 and RF12. Figure 7 In the example shown, the target trajectory TR11 is drawn along the center of the driving lane so as not to interfere with the risk potential field RF11. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR11. Figure 8In the example shown, a target trajectory TR12 is generated that bypasses the risk potential field RF12 that extends to near the center of the driving lane. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR12.

[0124] 3-2. Second embodiment

[0125] Figure 9 and Figure 10 1 is a conceptual diagram for explaining a second embodiment of the driving assistance control performed by the driving assistance system 100. The second embodiment of the driving assistance control is an embodiment of potential risk avoidance control. Figure 9 and Figure 10 In each of the above, the view from above of the driving lane is Figure 1 The diagram depicts the view forward from inside vehicle VH. In the second embodiment, parked vehicles PVL and PVS are located in the roadside strip between the left dividing line CL1 and the outer barrier wall BW. When viewed from vehicle VH traveling in the driving lane, a blind spot appears behind the parked vehicles PVL and PVS. The driving assistance system 100 identifies the parked vehicles PVL and PVS, located in front of vehicle VH and forming a blind spot when viewed from vehicle VH, as potential risk objects PR21 and PR22.

[0126] In potential risk avoidance control, it is assumed that there is a virtual pedestrian VP in the blind spot. Figure 9 In the example shown, the parking vehicle PVL is a large vehicle, so even if there is a pedestrian behind the parking vehicle PVL, the pedestrian will be completely hidden in the blind spot. Figure 9 In the example shown, the assumption that a virtual pedestrian VP exists in the blind spot of the parked vehicle PVL is maintained. Figure 10 In the example shown, the parking vehicle PVS is a small vehicle, so the pedestrian RP actually existing behind the parking vehicle PVS is not completely hidden in the blind spot of the parking vehicle PVS. The driving assistance system 100 can recognize the actual pedestrian RP as an object. Figure 10 In the case of the example shown, the assumption that a virtual pedestrian VP exists in the blind spot of the parked vehicle PVS is replaced by the fact that an actual pedestrian RP exists behind the parked vehicle PVS.

[0127] exist Figure 9 In the example shown, the driving assistance system 100 extracts object information related to the parked vehicle PVL as the potential risk object PR21 from the surrounding situation information as risk object information. Figure 10In the example shown, the driving assistance system 100 extracts object information related to the parked vehicle PVS, which is a potential risk object PR22, from the surrounding situation information as risk object information. It should be noted that the parked vehicles PVL and PVS are both potential risk objects that create blind spots and are themselves visible risk objects.

[0128] In the second embodiment, the driving assistance system 100 obtains information related to the moving object behind the potential risk objects PR21 and PR22 as the influencing factor information. Figure 9 In the example shown, the back of the parked vehicle PVL, which is the potential risk object PR21, is completely in the blind spot, so the presence of a moving object is unclear. The driving assistance system 100 obtains the fact that it is unclear whether there is a moving object behind the potential risk object PR21 as the influencing factor information. Figure 10 In the example shown, a pedestrian RP is detected behind the parked vehicle PVS as a potential risk object PR22. Figure 10 In the example shown, the influencing factor IF22 that affects the collision risk caused by the potential risk object PR22 is the actual pedestrian RP. The driving assistance system 100 obtains object information related to the actual pedestrian RP as the influencing factor information.

[0129] The driving assistance system 100 generates risk potential fields RF21 and RF22 based on risk object information and influencing factor information. Figure 9 In the example shown, it's unclear whether there's a moving object behind the potential risk object PR21. Therefore, the driving assistance system 100 generates a risk potential field RF21 centered around a virtual pedestrian VP, presumed to be in the blind spot. In this case, the driving assistance system 100 generates the risk potential field RF21 with a standard size determined based on the risk object information of the potential risk object PR21. Furthermore, since the parked vehicle PVL, representing the potential risk object PR21, is also a visible risk object, the driving assistance system 100 also generates a risk potential field RF210 extending around the parked vehicle PVL.

[0130] exist Figure 10In the example shown, the driving assistance system 100 generates a risk potential field RF22, which expands around the actual pedestrian RP, instead of the virtual pedestrian VP. Compared to the risk potential field RF21, which is set to predict the presence of a pedestrian, the risk potential field RF22 is set larger to avoid collision with the actual pedestrian RP. If the object information of the actual pedestrian RP includes a direction and speed of movement, the direction and extent of the expansion of the risk potential field RF22 can also be determined based on the direction and speed of movement. Furthermore, the parked vehicle PVS, a potential risk object PR22, is also a visible risk object. Therefore, the driving assistance system 100 also generates a risk potential field RF220 that expands around the parked vehicle PVS.

[0131] exist Figure 9 In the illustrated example, the driving assistance system 100 generates a target trajectory TR21 for the vehicle VH based on the risk potential fields RF21 and RF210. Specifically, the target trajectory TR21 is generated so as not to interfere with the risk potential field RF210 set around the parked vehicle PVL and the risk potential field RF21 set centered around the virtual pedestrian VP. The driving assistance system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR21.

[0132] exist Figure 10 In the illustrated example, the driving assistance system 100 generates a target trajectory TR22 for the vehicle VH based on the risk potential fields RF22 and RF220. Specifically, the target trajectory TR22 is generated so as not to interfere with the risk potential field RF220 set around the parked vehicle PVS and the risk potential field RF22 set centered on the actual pedestrian RP. Because the risk potential field RF22 extends to near the center of the driving lane, the target trajectory TR22 is generated to bypass the risk potential field RF22. The driving assistance system 100 determines the operating amount of each actuator so that the vehicle VH follows the target trajectory TR22.

[0133] 3-3. Third embodiment

[0134] Figure 11 and Figure 12 This is a conceptual diagram illustrating a third embodiment of driving assistance control performed by the driving assistance system 100. The third embodiment of driving assistance control is an embodiment of potential risk avoidance control. In the third embodiment, a parked vehicle PV is located in the roadside strip between the left dividing line CL1 and the outer barrier wall BW. When viewed from a vehicle VH traveling in the driving lane, a blind spot appears behind the parked vehicle PV. The driving assistance system 100 identifies the parked vehicle PV, located in front of the vehicle VH and forming an area that constitutes a blind spot when viewed from the vehicle VH, as potential risk objects PR31 and PR32.

[0135] In the third embodiment, the driving assistance system 100 extracts object information related to the parked vehicles PV, which are potential risk objects PR31 and PR32, from the surrounding situation information as risk object information. It should be noted that the parked vehicles PV are both potential risk objects that create blind spots and are themselves visible risk objects.

[0136] In potential risk avoidance control, such as Figure 11 and Figure 12 As shown in FIG, it is assumed that there is a virtual pedestrian VP in the blind spot formed by the potential risk objects PR31 and PR32. However, this assumption is based on a weak possibility that there may be a pedestrian. Figure 12 As in the example shown, if a pedestrian RP is standing on the opposite side of the travel lane and is waving behind the parked vehicle PV, there is a high probability that someone is at the destination of the gesture. In other words, there is a high probability that a pedestrian is in the blind spot formed by the potential risk object PR32.

[0137] In the third embodiment, the driving assistance system 100 obtains information related to dynamic factors acting on the blind spots formed by the potential risk objects PR31 and PR32 as influencing factor information. Figure 11 In the example shown, there is nothing around the parked vehicle PV, which is the potential risk object PR31. Therefore, there are no dynamic factors that act on the blind spot formed by the potential risk object PR31. The driving assistance system 100 obtains the fact that there are no dynamic factors that act on the blind spot formed by the potential risk object PR31 as influencing factor information. Figure 12 In the example shown, a pedestrian RP is identified as a dynamic factor acting on the blind spot formed by the potential risk object PR31. Figure 12 In the example shown, the influencing factor IF32 that influences the collision risk posed by the potential risk object PR32 is the actual pedestrian RP. More specifically, the actual pedestrian RP, who signals toward the blind spot formed by the potential risk object PR31, is the influencing factor IF32. The driving assistance system 100 obtains object information related to the actual pedestrian RP as influencing factor information.

[0138] The driving assistance system 100 generates risk potential fields RF31 and RF32 that extend around the virtual pedestrian VP based on the risk object information and the influencing factor information. Figure 11In the illustrated example, there are no dynamic factors acting on the blind spot formed by the potential risk object PR31. Therefore, the driving assistance system 100 generates a risk potential field RF31 of a standard size determined based on the risk object information of the potential risk object PR31. Furthermore, the parked vehicle PV, representing the potential risk object PR31, is also a visible risk object. Therefore, the driving assistance system 100 also generates a risk potential field RF310 extending around the parked vehicle PVL.

[0139] exist Figure 12 In the example shown, information related to the actual pedestrian RP acting in the blind spot formed by the potential risk object PR32 is used as influencing factor information. Since the actual pedestrian RP is signaling toward the blind spot, there is a high probability that a pedestrian is actually present in the blind spot. The driving assistance system 100 generates a risk potential field RF32 that further expands in the direction of the actual pedestrian RP compared to the risk potential field RF31 set as a possible pedestrian. In addition, the parked vehicle PV, which is the potential risk object PR32, is also a clear risk object, so the driving assistance system 100 also generates a risk potential field RF320 that expands around the parked vehicle PV. Moreover, the actual pedestrian RP itself is also a clear risk object, so the driving assistance system 100 also generates a risk potential field RF321 that expands around the actual pedestrian RP.

[0140] exist Figure 11 In the example shown, the driving assistance system 100 generates a target trajectory TR31 of the vehicle VH based on the risk potential fields RF31 and RF310. Figure 11 In the example shown, a target trajectory TR31 is generated so as not to interfere with the risk potential field RF310 set around the parked vehicle PV and the risk potential field RF31 set centered on the virtual pedestrian VP. The driving assistance system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR31.

[0141] exist Figure 12 In the example shown, the driving assistance system 100 generates a target trajectory TR32 of the vehicle VH based on the risk potential fields RF32, RF320, and RF321. Figure 12 In the example shown, a target trajectory TR32 is generated so as to pass between a risk potential field RF32 set around a virtual pedestrian VP and a risk potential field RF321 set around an actual pedestrian RP. The driving assistance system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR32.

[0142] 3-4. Fourth embodiment

[0143] Figure 13 and Figure 14 This is a conceptual diagram for explaining the fourth embodiment of the driving assistance control performed by the driving assistance system 100. The fourth embodiment of the driving assistance control is an embodiment of potential risk avoidance control. In the fourth embodiment, a branch road is connected to the right side of the driving lane defined by two dividing lines CL1 and CL2. A retaining wall BW is erected on the right side of the driving lane and on both sides of the branch road. Therefore, when viewed from the vehicle VH traveling in the driving lane, a blind spot formed by the retaining wall BW appears at the corner where the branch road connects to the driving lane. The driving assistance system 100 identifies the retaining wall BW that exists in front of the vehicle VH and forms an area that becomes a blind spot when viewed from the vehicle VH as potential risk markers PR41 and PR42.

[0144] In the potential risk avoidance control, it is assumed that a virtual pedestrian VP exists in the blind spot of the potential risk objects PR41 and PR42. The driving assistance system 100 extracts risk object information related to the virtual pedestrian VP from the surrounding situation information and obtains influencing factor information related to the virtual pedestrian VP. The risk object information is object information related to the barrier wall BW as the potential risk objects PR41 and PR42. Figure 13 The examples shown and Figure 14 The examples shown are universal. On the other hand, the influencing factor information is Figure 13 The examples shown and Figure 14 The example shown is different.

[0145] In the fourth embodiment, the driving assistance system 100 obtains information related to the time and location of detection of potential risk objects PR41 and PR42 as influencing factor information. The magnitude of the collision risk posed by potential risk objects PR41 and PR42 is related to both time and location. More specifically, the combination of time and location influences the magnitude of the collision risk posed by potential risk objects PR41 and PR42.

[0146] exist Figure 13 The examples shown and Figure 14 In the example shown, the factors IF411 and IF421 affecting the location of the collision risk are both schools. However, the factors IF412 and IF422 affecting the time of the collision risk are both Figure 13 The examples shown and Figure 14 The example shown is different. Figure 13 In the example shown, the time of the influencing factor IF412 is 10 o'clock outside the school hours. Figure 14 In the example shown, the time of the influencing factor IF422 is 8 o'clock during the school rush hour. During the school rush hour, many children walk around the school, so the risk of children running out during the school rush hour will inevitably increase. Figure 14In the example shown, the risk ratio of the hypothetical pedestrian VP to run away is Figure 13 The risk of the imaginary pedestrian VP running away is high in the example shown.

[0147] The driving assistance system 100 generates risk potential fields RF41 and RF42 extending around the virtual pedestrian VP based on the risk object information and influencing factor information related to the virtual pedestrian VP. Figure 13 In the example shown, the location is a school and the time is 10 o'clock outside of the school hours as the influencing factor information. Figure 14 In the example shown, the location is a school and the time is 8:00 during the school commute time. The location-related information in the influencing factor information can be obtained from the map information, and the time-related information can be obtained from the built-in clock of the control device 20.

[0148] In the fourth embodiment, the driving assistance system 100 sets the risk potential fields RF41 and RF42 as ellipses that expand from the branch road toward the driving lane. Figure 14 In the example shown, Figure 13 Compared to the risk potential field RF41 in the example shown, the driving assistance system 100 further expands the risk potential field RF42 from the junction toward the driving lane.

[0149] The driving assistance system 100 generates target trajectories TR41 and TR42 of the vehicle VH based on the risk potential fields RF41 and RF42. Figure 13 In the example shown, the target trajectory TR41 is drawn along the center of the driving lane so as not to interfere with the risk potential field RF41. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR41. Figure 14 In the example shown, a target trajectory TR42 is generated that bypasses the risk potential field RF42 that extends to near the center of the driving lane. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR42.

[0150] 3-5. Fifth embodiment

[0151] Figures 15 to 17 This is a conceptual diagram illustrating a fifth embodiment of driving assistance control performed by the driving assistance system 100. The fifth embodiment of driving assistance control is an embodiment of explicit risk avoidance control. In the fifth embodiment, a parked vehicle PV is located in the roadside strip between the left dividing line CL1 and the outer barrier wall BW. The driving assistance system 100 identifies the parked vehicle PV, which is located in front of the vehicle VH and has the potential to collide with the vehicle VH, as explicit risk objects ER51, ER52, and ER53.

[0152] In the fifth embodiment, the driving assistance system 100 extracts object information related to the parked vehicles PV, which are explicit risk objects ER51, ER52, and ER53, from the surrounding situation information as risk object information. It should be noted that the parked vehicles PV are explicit risk objects that could collide with the vehicle VH and are also potential risk objects that create a blind spot when viewed from the vehicle VH.

[0153] The collision risk posed by a parked vehicle PV refers to the risk of a collision with vehicle VH due to the parked vehicle PV starting to move. The probability of the parked vehicle PV starting to move is higher in the case where the parked vehicle PV is unoccupied or where a driver is seated inside the parked vehicle PV. Failure to detect a driver does not necessarily mean the parked vehicle PV is unoccupied, but the probability of the parked vehicle PV starting to move is higher when a driver is actually detected. It should be noted that the driver inside the parked vehicle PV can be detected from camera images.

[0154] Furthermore, when a driver is seated in a parked vehicle (PV), the likelihood of the parked vehicle (PV) starting to move is higher when the driver is not performing any operation or when the driver is performing some operation. The activation of the brake lights is a driver operation that can be detected visually. The likelihood of the parked vehicle (PV) starting to move is higher when the brake lights are detected than when they are not. It should be noted that the activation of the brake lights can be detected from camera images.

[0155] In the fifth embodiment, the driving assistance system 100 obtains information related to the presence or absence of a driver in the parking vehicle PV displayed as risk objects ER51, ER52, and ER53 as influencing factor information. In addition, the driving assistance system 100 obtains information related to the presence or absence of the brake lights of the parking vehicle PV as influencing factor information. Figure 15 In the example shown, no driver is detected in the parked vehicle PV, and no brake lights are detected to be on. The driving assistance system 100 obtains the fact that no driver is detected in the parked vehicle PV and the brake lights are not on as influencing factor information.

[0156] exist Figure 16 In the example shown, the driver DV is detected in the parked vehicle PV. However, the lighting of the brake lights is not detected. Figure 16 In the example shown, the presence of the driver DV becomes the influencing factor IF52 that affects the collision risk caused by the parked vehicle PV. The driving assistance system 100 obtains the fact that the driver DV is inside the parked vehicle PV as influencing factor information.

[0157] exist Figure 17 In the example shown, the driver DV is detected in the parked vehicle PV. Furthermore, the lighting of the brake lights BL is also detected. Figure 17 In the example shown, the presence of the driver DV serves as an influencing factor IF531 affecting the collision risk posed by the parked vehicle PV. Furthermore, the illumination of the brake lights BL serves as an influencing factor IF532 affecting the collision risk posed by the parked vehicle PV. The driving assistance system 100 obtains the presence of the driver DV and the illumination of the brake lights BL within the parked vehicle PV as influencing factor information.

[0158] The driving assistance system 100 generates risk potential fields RF51, RF52, and RF53 that extend around the parked vehicle PV, which is a visible risk object ER51, ER52, and ER53, based on the risk object information and the influencing factor information. Figure 15 In the illustrated example, there are no factors that increase the likelihood that the parked vehicle PV will begin to move. Therefore, the driving assistance system 100 generates a risk potential field RF51 of a standard size determined based on the risk object information of the parked vehicle PV. Furthermore, the parked vehicle PV, which is a visible risk object ER51, is also a potential risk object. Therefore, the driving assistance system 100 also generates a risk potential field RF510 that extends from a hypothetical pedestrian VP, who is assumed to be in the blind spot of the parked vehicle PV.

[0159] exist Figure 16 In the example shown, the driver DV is riding in the parked vehicle PV, so the probability that the parked vehicle PV starts to move is Figure 15 The driving assistance system 100 generates the same Figure 15 The risk potential field RF52 is further expanded compared to the risk potential field RF51 set in the illustrated example. Furthermore, the driving assistance system 100 also generates a risk potential field RF520 that expands around the virtual pedestrian VP in the blind spot of the parked vehicle PV.

[0160] exist Figure 17 In the example shown, the driver DV is riding in the parked vehicle PV and the brake lights BL are also on, so the probability that the parked vehicle PV starts moving is Figure 16 The driving assistance system 100 generates the same Figure 16 In the example shown, the risk potential field RF53 is further expanded compared to the risk potential field RF52 set. In addition, the driving assistance system 100 also generates a risk potential field that expands around the virtual pedestrian VP in the blind spot of the parked vehicle PV. Figure 17In the example shown, the risk potential field centered on the virtual pedestrian VP is covered by the large risk potential field RF53 generated around the parked vehicle PV.

[0161] exist Figure 15 In the example shown, the driving assistance system 100 generates a target trajectory TR31 of the vehicle VH based on the risk potential fields RF51 and RF510. Figure 15 In the example shown, a target trajectory TR51 is generated so as not to interfere with the risk potential field RF51 set around the parked vehicle PV and the risk potential field RF510 set centered on the virtual pedestrian VP. The driving assistance system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR51.

[0162] exist Figure 16 In the example shown, the driving assistance system 100 generates a target trajectory TR52 of the vehicle VH based on the risk potential fields RF52 and RF520. Figure 16 In the example shown, the target trajectory TR52 is generated so as to bypass the expanded risk potential field RF52. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR52.

[0163] exist Figure 17 In the example shown, the driving assistance system 100 generates a target trajectory TR53 of the vehicle VH based on the risk potential field RF53. Figure 17 In the example shown, the target trajectory TR53 is generated so as to significantly bypass the further expanded risk potential field RF53. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR53.

[0164] 3-6. Sixth embodiment

[0165] Figure 18 and Figure 19 This is a conceptual diagram illustrating a sixth embodiment of driving assistance control performed by the driving assistance system 100. This sixth embodiment of driving assistance control is an embodiment of explicit risk avoidance control. In this sixth embodiment, a pedestrian RP is actually present outside the left dividing line CL1. The driving assistance system 100 identifies the pedestrian RP, who is in front of the vehicle VH and has a potential collision risk with the vehicle VH, as explicit risk objects ER61 and ER62.

[0166] In the sixth embodiment, the driving assistance system 100 extracts object information related to pedestrians RP, which are the apparent risk objects ER61 and ER62 , from the surrounding situation information as risk object information.

[0167] The collision risk caused by pedestrians RP is affected by the state of the road on which the pedestrians RP are located. For example, if the roadway and the sidewalk are separated by guardrails, curbs, poles, etc., the collision risk caused by pedestrians RP on the sidewalk is reduced compared to a case without such structures. Figure 19 As in the example shown, at a road construction site surrounded by a plurality of road cones RC, the possibility of a pedestrian RP stepping outside the road cones RC is low, and thus the collision risk caused by the pedestrian RP is reduced.

[0168] In the sixth embodiment, the driving assistance system 100 obtains information related to the state of the road where the risk objects ER61 and ER62 are detected as the influencing factor information. Figure 18 In the example shown, there are no objects around the pedestrian RP, which is the risk object ER61, that would prevent the pedestrian RP from entering the lane. The driving assistance system 100 obtains the fact that there are no objects that would prevent the pedestrian RP from moving freely in the road where the pedestrian RP is detected as influencing factor information.

[0169] exist Figure 19 In the example shown, the pedestrian RP is at a road construction site, and a plurality of road cones RC are detected around the pedestrian RP, separating the construction site from the lane. Figure 19 In the example shown, the fact that the road on which the pedestrian RP is standing is under construction constitutes an influencing factor IF62 that affects the collision risk posed by the pedestrian RP. The driving assistance system 100 obtains the fact that the road on which the pedestrian RP was detected is under construction as influencing factor information. Information related to influencing factor IF62 can be obtained, for example, from camera images or from road traffic information transmitted from a road traffic information system.

[0170] The driving assistance system 100 generates risk potential fields RF61 and RF62 that extend around pedestrians RP, which are explicit risk objects ER61 and ER62, based on the risk object information and the influencing factor information. Figure 18 In the example shown, there are no factors that hinder the movement of the pedestrian RP, so the driving assistance system 100 generates a risk potential field RF61 of a standard size determined based on the risk object information of the pedestrian RP.

[0171] exist Figure 19 In the example shown, a pedestrian RP is standing at a road construction site surrounded by a plurality of traffic cones RC. The possibility that a pedestrian RP, i.e., an operator, will cross the traffic cones RC and move toward the lane during the road construction is low. Figure 18Compared with the example shown in FIG, the possibility of the pedestrian RP rushing out in the direction of the driving lane is low, and the collision risk caused by the pedestrian RP is also low. Figure 18 In the example shown, the risk potential field RF61 is set to be a risk potential field RF62 that is further reduced.

[0172] The driving assistance system 100 generates target trajectories TR61 and TR62 of the vehicle VH based on the risk potential fields RF61 and RF62. Figure 18 In the example shown, a target trajectory TR61 is generated that circumvents the risk potential field RF61 that extends to the driving lane. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR61. Figure 19 In the example shown, the risk potential field RF62 is limited to the area surrounded by the plurality of road cones RC, so the target trajectory TR62 is generated along the center of the driving lane. The driving assistance system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR62.

[0173] 3-7. Seventh embodiment

[0174] Figures 20 to 22 This is a conceptual diagram illustrating a seventh embodiment of driving assistance control performed by the driving assistance system 100. This seventh embodiment of driving assistance control is an embodiment of explicit risk avoidance control. In this seventh embodiment, a pedestrian RP is present on an undivided road. The driving assistance system 100 identifies the pedestrian RP, who is in front of the vehicle VH and has a potential collision risk with the vehicle VH, as explicit risk objects ER71, ER72, and ER73.

[0175] In the seventh embodiment, the driving assistance system 100 extracts object information related to pedestrians RP, which are the apparent risk objects ER71 , ER72 , and ER73 , from the surrounding situation information as risk object information.

[0176] The collision risk caused by a pedestrian RP is affected by the time and place when the pedestrian RP is detected. In more detail, the combination of time and place will affect the collision risk caused by the pedestrian RP. For example, for a situation where a pedestrian RP is walking in a place that is not a busy street during the day and for a situation where the pedestrian RP is walking in a busy street at night, the risk of a collision between the vehicle VH and the pedestrian RP in the latter situation is higher. This is because the pedestrian RP may have been drinking in the latter situation. Moreover, by combining the movement of the pedestrian RP with the combination of time and place, the accuracy of the collision risk estimation can be further improved. For example, for a pedestrian RP walking straight in a busy street at night and a pedestrian RP walking while swaying, the latter is more likely to be drunk and thus has a higher collision risk.

[0177] In the seventh embodiment, the driving assistance system 100 acquires information related to the time and location of pedestrians RP detected as risk objects ER71, ER72, and ER73 as influencing factor information. Furthermore, the driving assistance system 100 acquires information related to the pedestrians RP's past location history as influencing factor information. This past location history can be used to determine whether the pedestrians RP are walking straight or swaying. Location information within the influencing factor information can be acquired from map information, while time information can be acquired from the built-in clock of the control device 20. The pedestrians RP's past location history can be acquired from the pedestrians RP's object information.

[0178] exist Figure 20 In the example shown, a pedestrian RP, representing a significant risk object ER71, is walking straight in a non-bustling area during the day. The driving assistance system 100 obtains as influencing factor information the following: the pedestrian RP was detected during the day, the pedestrian was detected in a non-bustling area, and the pedestrian was walking straight.

[0179] exist Figure 21 In the example shown, a pedestrian RP, representing a significant risk object ER72, is walking straight in a busy street at night. Nighttime, as the time of day, is detected as an influencing factor IF721 affecting the collision risk posed by the pedestrian RP. Furthermore, a busy street, as the location, is also detected as an influencing factor IF722 affecting the collision risk posed by the pedestrian RP. The driving assistance system 100 obtains as influencing factor information the following: the time of day when the pedestrian RP was detected, the location where the pedestrian RP was detected being a busy street, and the fact that the pedestrian RP was walking straight.

[0180] exist Figure 22 In the example shown, a pedestrian RP, which is a visible risk object ER73, is walking swaying in a busy street at night. Nighttime, which is the time, is detected as an influencing factor IF731 that affects the collision risk generated by the pedestrian RP. In addition, a busy street, which is the location, is also detected as an influencing factor IF732 that affects the collision risk generated by the pedestrian RP. Moreover, shaking, which is the location history, is also detected as an influencing factor IF733 that affects the collision risk generated by the pedestrian RP. The driving assistance system 100 obtains, as influencing factor information, the fact that the time when the pedestrian RP was detected was nighttime, the location where the pedestrian RP was detected was a busy street, and that the pedestrian RP was swaying.

[0181] The driving assistance system 100 generates risk potential fields RF71, RF72, and RF73 that extend around pedestrians RP, which are explicit risk objects ER71, ER72, and ER73, based on the risk object information and the influencing factor information. Figure 20 In the example shown, there is no influencing factor indicating the possibility that the pedestrian RP is drinking alcohol. Therefore, the driving assistance system 100 generates a risk potential field RF71 of a standard size determined based on the risk object information of the pedestrian RP.

[0182] exist Figure 21 In the example shown, the pedestrian RP is walking in a busy street at night, so the pedestrian RP is more likely to drink alcohol than Figure 20 The pedestrian RP in the example shown is likely to have been drinking. If the pedestrian RP is drinking, he or she may not notice the vehicle VH due to a decrease in judgment. Therefore, the higher the probability that the pedestrian RP is drinking, the greater the risk of collision. Therefore, the driving assistance system 100 generates the same Figure 20 The risk potential field RF71 set in the illustrated example is a risk potential field RF72 that is further expanded in all directions compared to the risk potential field RF71.

[0183] exist Figure 22 In the example shown, the pedestrian RP is walking in a busy street at night and is wobbling. Therefore, the probability that the pedestrian RP is drunk is higher than Figure 21 The pedestrian RP in the example shown is more likely to be drunk. If the pedestrian RP is drunk to the point of being unsteady on his feet, the pedestrian RP may make unpredictable movements. Therefore, if the pedestrian RP is more likely to be drunk, the risk of collision becomes greater. Therefore, the driving assistance system 100 generates the same Figure 21 The risk potential field RF72 set in the illustrated example is a risk potential field RF73 that is further expanded in all directions compared to the risk potential field RF72.

[0184] The driving assistance system 100 generates target trajectories TR71, TR72, TR73 of the vehicle VH based on the risk potential fields RF71, RF72, and R73. Figure 20 In the example shown, the target trajectory TR71 is generated so as not to interfere with the risk potential field RF71 set around the pedestrian RP. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR71. Figure 21 In the example shown, the target trajectory TR72 is generated so as to bypass the expanded risk potential field RF72. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR72. Figure 22In the example shown, the target trajectory TR73 is generated so as to significantly bypass the further expanded risk potential field RF73. The driving support system 100 determines the operation amount of each actuator so that the vehicle VH follows the target trajectory TR73.

[0185] 4. Other Implementation Methods

[0186] When determining a risk value that quantifies the collision risk, the risk field disclosed in Patent Document 1 (Japanese Patent Application Laid-Open No. 2017-206117) may be calculated instead of the risk potential field.

[0187] The above-mentioned embodiments of potential risk avoidance control can be implemented in combination as appropriate. The above-mentioned embodiments of explicit risk avoidance control can be implemented in combination as appropriate. Furthermore, the above-mentioned embodiments of potential risk avoidance control can be implemented in combination with the above-mentioned embodiments of explicit risk avoidance control as appropriate.

Claims

1. A driving assistance system for assisting the driving of a vehicle, wherein the driving assistance system comprises: at least one memory storing at least one program; and at least one processor coupled to the at least one memory, The at least one program is configured to cause the at least one processor to execute: extracting, from information about a surrounding condition of the vehicle, information related to a potential risk object that exists in front of the vehicle and forms a blind spot when viewed from the vehicle as risk object information related to a risk object that poses a collision risk to the vehicle; Acquiring influencing factor information related to an influencing factor, where the influencing factor is a factor that exists independently of the potential risk object and affects the collision risk; determining a risk value that quantifies the collision risk based on the risk object information and the influencing factor information, wherein the risk value is given as a distribution in a vehicle coordinate system or an absolute coordinate system; generating a risk potential field representing a distribution of the risk values; plotting a target trajectory of the vehicle in a manner that does not interfere with the risk potential field; as well as Based on the determined risk value, an operation amount of an actuator that controls the motion of the vehicle is determined so that the vehicle follows the target trajectory.

2. The driving assistance system according to claim 1, characterized in that: The at least one program is configured to enable the at least one processor to execute: obtaining information related to the surrounding environment of the potential risk object as the influencing factor information.

3. The driving assistance system according to claim 1, characterized in that: The at least one program is configured to cause the at least one processor to execute: obtaining information related to a moving object behind the potential risk object as the influencing factor information.

4. The driving assistance system according to claim 1, characterized in that: The at least one program is configured to cause the at least one processor to execute: obtaining information related to dynamic factors acting on the blind spot formed by the potential risk object as the influencing factor information.

5. The driving assistance system according to claim 1, characterized in that: The at least one program is configured to cause the at least one processor to execute: obtaining information related to the time and location of detecting the potential risk object as the influencing factor information.

6. A driving assistance method for assisting driving of a vehicle, the driving assistance method comprising: extracting, from information about a surrounding condition of the vehicle, information related to a potential risk object that exists in front of the vehicle and forms a blind spot when viewed from the vehicle as risk object information related to a risk object that poses a collision risk to the vehicle; Acquiring influencing factor information related to an influencing factor, where the influencing factor is a factor that exists independently of the potential risk object and affects the collision risk; determining a risk value that quantifies the collision risk based on the risk object information and the influencing factor information, wherein the risk value is given as a distribution in a vehicle coordinate system or an absolute coordinate system; generating a risk potential field representing a distribution of the risk values; plotting a target trajectory of the vehicle in a manner that does not interfere with the risk potential field; as well as Based on the determined risk value, an operation amount of an actuator that controls the motion of the vehicle is determined so that the vehicle follows the target trajectory.

7. A computer-readable recording medium having recorded thereon a program configured to cause a processor to execute a process, wherein the computer-readable recording medium is characterized in that: The processing includes: extracting, from information about a surrounding condition of a vehicle, information related to a potential risk object that exists in front of the vehicle and forms an area that becomes a blind spot when viewed from the vehicle as risk object information related to a risk object that poses a risk of collision to the vehicle; Acquiring influencing factor information related to an influencing factor, where the influencing factor is a factor that exists independently of the potential risk object and affects the collision risk; determining a risk value that quantifies the collision risk based on the risk object information and the influencing factor information, wherein the risk value is given as a distribution in a vehicle coordinate system or an absolute coordinate system; generating a risk potential field representing a distribution of the risk values; plotting a target trajectory of the vehicle in a manner that does not interfere with the risk potential field; as well as Based on the determined risk value, an operation amount of an actuator that controls the motion of the vehicle is determined so that the vehicle follows the target trajectory.

Images