Drive support system

Abstract

To provide a drive support system capable of vehicle control of automatic drive or drive assist reflecting a vehicle control behavior at the time of manual drive by a driver.SOLUTION: In a drive support system, feature information of a driver of an own vehicle and travel data representing past vehicle control behavior and a surrounding environment at the time of manual drive of the driver of the own vehicle are input to a learned parameter setting model, and a control parameter output from the learned parameter setting model is set as a control parameter used in a vehicle controller that performs automatic drive or drive support.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] This disclosure relates to driver assistance systems. [Background technology]

[0002] In recent years, with the aim of ensuring safe vehicle driving, driving assistance systems that assist the driver in driving operations and automated driving systems in which the vehicle performs driving operations have been developed. Conventionally, systems have been developed to ensure safe vehicle driving, but they have not provided vehicle control that provides automated driving or driving assistance while taking into account the driver's vehicle control preferences.

[0003] In Patent Document 1, the system is configured to take into consideration the driver's preferences, such as prioritizing "fuel efficiency" or "driving performance," and select a learned control model that reflects the driver's past driving information and changes engine control variables such as ignition timing, fuel injection amount, injection timing, throttle opening, variable valve timing, and EGR valve, so as to reflect the driver's preferences as much as possible, within a range that minimizes the output parameters of "NOx emissions" and "consumption required for restarting after idle stop." [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Publication No. 2022-67454 Summary of the Invention [Problem to be solved by the invention]

[0005] However, the technology of Patent Document 1 is a technology for selecting a learned engine control model that takes into account the driver's preferences, and cannot be applied to a vehicle controller that performs automatic driving or driving assistance.

[0006] Therefore, an object of the present disclosure is to provide a driving assistance system that can perform vehicle control for automatic driving or driving assistance that reflects the vehicle control behavior during manual driving by the driver. [Means for solving the problem]

[0007] The driving assistance system according to the present disclosure includes: an information acquisition unit that acquires the running state of the host vehicle and the surrounding environment of the host vehicle; a vehicle control unit that uses a vehicle controller that performs automatic driving or driving assistance and that is capable of changing a control constant that affects a control behavior of the host vehicle based on the running state of the host vehicle and the surrounding environment, calculates a target value for vehicle control of the host vehicle, and controls the host vehicle based on the target value; a control constant setting unit that uses characteristic information of a target driver and driving data representing the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, and outputs control constants that will result in vehicle control behavior that corresponds to the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, inputs the characteristic information of the driver of the vehicle and the driving data representing the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, and sets the control constants output from the learned constant setting model as the control constants used in the vehicle controller; It is equipped with the following. [Effects of the Invention]

[0008] According to the driving assistance system of the present disclosure, a learned constant setting model is used to set control constants that result in vehicle control behavior that corresponds to characteristic information of the driver of the vehicle and past driving data when the driver of the vehicle is manually driving, and automatic driving control or driving assistance control of the vehicle is performed using a vehicle controller having the set control constants. Thus, automatic driving control or driving assistance control that corresponds to manual driving by the driver of the vehicle can be performed, and the ride comfort of the driver of the vehicle can be improved. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a schematic block diagram of a driving assistance device according to a first embodiment. [Figure 2] 1 is a schematic hardware configuration diagram of a driving assistance device according to a first embodiment. [Figure 3] FIG. 3 is a diagram for explaining learning of a learned constant setting model according to the first embodiment. [Figure 4] FIG. 3 is a diagram for explaining the setting of control constants using a learned constant setting model according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0010] 1. First Embodiment A driving assistance system according to embodiment 1 will be described with reference to the drawings. In this embodiment, the driving assistance system is incorporated into a driving assistance device 50 provided in the vehicle.

[0011] As shown in FIG. 1, the vehicle is equipped with a surroundings monitoring device 31, a position detection device 32, a vehicle state detection device 33, a map information database 34, a wireless communication device 35, a driving assistance device 50, a drive control device 36, a power plant 8, an electric steering device 7, an electric braking device 9, a human interface device 37, and an occupant monitoring device 38, etc.

[0012] The periphery monitoring device 31 is a device such as a camera, radar, or ultrasonic sensor that monitors the periphery of the vehicle. The radar may be a millimeter wave radar, laser radar, or the like. The wireless communication device 35 performs wireless communication with a base station using a cellular wireless communication standard such as 4G or 5G. The wireless communication device 35 also performs wireless communication with roadside devices, surrounding vehicles, and the like.

[0013] The position detection device 32 is a device that detects the current position (latitude, longitude, altitude) of the vehicle, and uses a GPS antenna or the like that receives signals output from artificial satellites such as the GNSS (Global Navigation Satellite System).

[0014] The map information database 34 stores road information such as road shapes (e.g., the number of lanes, the position of each lane, the shape of each lane, the type of each lane, the road type, the speed limit, etc.), signs, traffic lights, etc. The shape of the lane includes the width, angle, curvature of the lane, etc. The map information database 34 is mainly composed of a storage device. The map information database 34 may be provided in a server outside the vehicle connected to a network, and the driving assistance device 50 may obtain necessary road information from the server outside the vehicle via the wireless communication device 35.

[0015] The drive control device 36 includes a power control device, a braking control device, an automatic steering control device, a light control device, etc. The power control device controls the output of a power machine 8 such as an internal combustion engine or a motor. The braking control device controls the braking force of an electric braking device 9. The automatic steering control device controls the electric steering device 7. The light control device controls turn signals, hazard lights, etc.

[0016] The vehicle state detection device 33 is a detection device that detects the running state of the host vehicle. In this embodiment, the vehicle state detection device 33 detects the speed, acceleration, yaw rate, lateral acceleration, steering angle, steering angular velocity, direction, etc. of the host vehicle as the running state of the host vehicle. For example, the vehicle state detection device 33 may be provided with a speed sensor, an acceleration sensor, an angular velocity sensor that detects the rotation speed of the wheels, a steering angle sensor, a direction sensor, etc.

[0017] The vehicle state detection device 33 detects the acceleration / deceleration operation, steering angle operation, and lane change operation by the driver as the running state of the vehicle. For example, an accelerator position sensor, a brake position sensor, a steering angle sensor (handle angle sensor), a steering torque sensor, a turn signal position switch, etc.

[0018] The human interface device 37 is a device that receives input from the driver through a speaker, a display screen, an input device, etc., and transmits information to the driver.

[0019] The occupant monitoring device 38 is a monitoring device that monitors the state of an occupant. A camera that captures an image of an occupant (for example, an occupant (driver) sitting in the driver's seat) is used as the occupant monitoring device 38. Alternatively, various types of biological information sensors that detect biological information of the occupant (for example, heart rate, blood pressure, respiratory rate) may be used as the occupant monitoring device 38.

[0020] 1-1. Driving assistance device 50 The driving assistance device 50 includes processing units such as an information acquisition unit 51, a control constant setting unit 52, a vehicle control unit 53, and an occupant state estimation unit 54. Each process of the driving assistance device 50 is realized by a processing circuit included in the driving assistance device 50. Specifically, as shown in Fig. 2, the driving assistance device 50 includes an arithmetic processing device 90 such as a CPU (Central Processing Unit), a storage device 91, an input / output device 92 that inputs and outputs external signals to the arithmetic processing device 90, and the like.

[0021] The arithmetic processing device 90 may be an ASIC (Application Specific Integrated Circuit), an IC (Integrated Circuit), a DSP (Digital Signal Processor), an FPGA (Field Programmable Gate Array), a GPU (Graphics Processing Unit), an AI (Artificial Intelligence) chip, various logic circuits, various signal processing circuits, etc. Furthermore, the arithmetic processing device 90 may be a plurality of the same or different types, and each process may be shared and executed. The storage device 91 may be a variety of storage devices, such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EEPROM (Electrically Erasable Programmable Read Only Memory), a hard disk, etc.

[0022] The input / output device 92 is equipped with a communication device, an A / D converter, an input / output port, a drive circuit, etc. The input / output device 92 is connected to the surroundings monitoring device 31, the position detection device 32, the vehicle state detection device 33, the map information database 34, the wireless communication device 35, the drive control device 36, the human interface device 37, the occupant monitoring device 38, etc., and communicates with these devices.

[0023] The processing of each of the processing units 51 to 54 provided in the driving assistance device 50 is realized by the arithmetic processing device 90 executing software (programs) stored in the storage device 91 and cooperating with other hardware of the driving assistance device 50, such as the storage device 91 and the input / output device 92. Note that setting data such as setting values ​​of the learned constant setting model used by each of the processing units 51 to 54 is stored in the storage device 91, such as an EEPROM.

[0024] 1-1-1. Information acquisition section 51 The information acquisition unit 51 acquires the traveling state of the host vehicle. In this embodiment, the information acquisition unit 51 acquires the position information of the host vehicle from the position detection device 32. Note that the position of the host vehicle may be detected by various methods such as a map matching method, a dead reckoning method, or a method using detected information around the host vehicle.

[0025] The information acquisition unit 51 acquires the speed, acceleration, jerk, yaw rate, steering angle, lateral acceleration, moving direction, etc. of the host vehicle from the vehicle state detection device 33 .

[0026] The information acquisition unit 51 acquires the driver's acceleration / deceleration operations (in this example, accelerator pedal operation and brake pedal operation), steering angle operation (steering angle and steering torque), and turn signal operation from the vehicle state detection device 33. In addition, the information acquisition unit 51 acquires the output of the power machine 8 and its target value, the braking force of the electric braking device 9 and its target value, the steering angle of the electric steering device 7 and its target value, and the operating state of the turn signal and its command from the vehicle control unit 53 or the drive control device 36.

[0027] The information acquisition unit 51 acquires information about the surrounding environment of the vehicle. The information acquisition unit 51 acquires road information about the surrounding area of ​​the vehicle from the map information database 34 based on the position information of the vehicle acquired from the position detection device 32. The acquired road information includes information such as the number of lanes, the position of each lane, the shape of each lane, the type of each lane, the road type, and the speed limit.

[0028] The information acquisition unit 51 acquires information about vehicles surrounding the vehicle based on the detection information acquired from the periphery monitoring device 31. The acquired information about the surrounding vehicles includes the position, moving direction, speed, and acceleration of the surrounding vehicles. In addition to the surrounding vehicles, the information acquisition unit 51 also acquires information about obstacles, pedestrians, signs, traffic regulations such as lane restrictions, and the like. The information acquisition unit 51 may acquire information about the surrounding environment, such as information about surrounding vehicles, road information, and obstacles, from surrounding vehicles and roadside devices.

[0029] The information acquisition unit 51 also detects the shape and type of road dividing lines and the like based on detection information of dividing lines such as white lines and road shoulders acquired from the periphery monitoring device 31, and determines the shape and position of each lane, the number of lanes, the type of each lane, etc., based on the detected shape and type of road dividing lines and the like. The information acquisition unit 51 also acquires road information such as speed limits, road types, and regulation information based on detection information of road signs, road markings, traffic guidance, etc. acquired from the periphery monitoring device 31.

[0030] 1-1-2. Vehicle control unit 53 The vehicle control unit 53 uses a vehicle controller that performs automatic driving or driving assistance and is capable of changing control constants that affect the control behavior of the vehicle based on the driving state of the vehicle and the surrounding environment acquired by the information acquisition unit 51, calculates target values ​​for vehicle control of the vehicle, and controls the vehicle based on the target values.

[0031] When performing autonomous driving or driving assistance, the vehicle control unit 53 determines a target driving state or a target driving trajectory based on the driving state of the host vehicle and the surrounding environment acquired by the information acquisition unit 51. The target driving state is a target value for various driving states, such as a target value for the host vehicle's speed, a target value for the distance between the vehicle in front and behind, and a target value for the host vehicle's lateral position. The type and setting method of the target value set as the target driving state are changed depending on the content of vehicle control, such as lane keeping control, speed control, distance control, obstacle avoidance control, lane change control, etc. The target driving trajectory is a driving plan of a time series of various target values, such as the position of the host vehicle, the traveling direction of the host vehicle, and the speed of the host vehicle at each future point in time. The type and setting method of the target value set as the target driving trajectory are changed depending on the content of vehicle control, such as lane keeping control, speed control, distance control, obstacle avoidance control, lane change control, etc.

[0032] In various vehicle controls such as lane keeping control, speed control, inter-vehicle distance control, obstacle avoidance control, and lane change control, maximum allowable acceleration, maximum allowable deceleration, target speed, target inter-vehicle distance, target lane change distance, target lane change trajectory, deceleration timing when entering a curve, and target entry speed are set, and these correspond to control constants. Note that the types and number of control constants can be changed as appropriate according to the types and number of control constants that can be set in the vehicle controller.

[0033] The maximum allowable acceleration places an upper limit on the maximum acceleration during acceleration. The characteristics of the acceleration control behavior are adjusted depending on the magnitude of the maximum allowable acceleration. The maximum allowable deceleration places an upper limit on the maximum deceleration during deceleration. The characteristics of the deceleration control behavior are adjusted depending on the magnitude of the maximum allowable deceleration. Note that the deceleration is the absolute value of the negative acceleration.

[0034] The target speed value is set for each road type or speed limit. The characteristics of the speed control behavior are adjusted depending on the magnitude of the target speed value. The target inter-vehicle distance value is set for each speed. The characteristics of the inter-vehicle distance control behavior are adjusted depending on the magnitude of the target inter-vehicle distance value. The target lane-change required distance and the target lane-change trajectory are set for each speed. The target lane-change required distance is the target value of the travel distance from the start to the completion of the lane change. The target lane-change trajectory is the target value of the lateral movement amount relative to the travel distance after the start of the lane change. The travel distance may be made dimensionless by dividing it by the target lane-change required distance. The characteristics of the lane-change control behavior are adjusted depending on the settings of these target values. The target deceleration timing and the entry speed when entering a curve are set for each curvature of the curve. The deceleration timing when entering a curve is the distance from the start of the curve to the point where deceleration begins. The entry speed is the speed at the start of the curve. The characteristics of the curve driving control behavior are adjusted according to these set values.

[0035] These control constants are set by a control constant setting unit 52, which will be described later. If the control constants are limited by limit values ​​by the control constant setting unit 52, the control constants after the control processing are set. The vehicle control unit 53 uses a vehicle controller in which the control constants have been set to calculate target values ​​for vehicle control so that the host vehicle follows a target driving state or target driving trajectory. Various types of controllers are used as the vehicle controller. Depending on the type of vehicle control, one or both of longitudinal control, which controls the behavior of the vehicle in the longitudinal direction, and lateral control, which controls the behavior of the vehicle in the lateral direction, are executed.

[0036] For example, a known feedback controller that performs feedback control or a known optimal controller that solves an optimization problem may be used as the vehicle controller.

[0037] When lateral control is performed, the target value for vehicle control is set to the steering angle. When longitudinal control is performed, the target value for vehicle control is set to the target value for acceleration of the host vehicle. The vehicle control unit 53 calculates the target value for the output of the power machine 8 and the target value for the braking force of the electric brake device 9 so that the acceleration of the host vehicle becomes the target value for acceleration.

[0038] <Manual operation> When the driver is driving manually, the vehicle control unit 53 calculates a target value for the output of the power machine 8 and a target value for the braking force of the electric braking device 9 based on the driver's operation of the accelerator pedal and the brake pedal. The vehicle control unit 53 calculates an operation command for the turn signal based on the driver's operation of the turn signal.

[0039] <Drive control> The vehicle control unit 53 transmits the target value of the output of the power machine 8, the target value of the braking force of the electric brake device 9, the target value of the steering angle, and operation commands for the turn signal lights to the drive control device 36. The drive control device 36 includes a power control device, a braking control device, an automatic steering control device, a light control device, etc.

[0040] The power control device controls the output of the power machine 8, such as an internal combustion engine or a motor, so that the output of the power machine 8 becomes a target output value. The braking control device controls the braking force of the electric braking device 9 so that the braking force of the electric braking device 9 becomes a target braking force value. When automatic driving or driving assistance is performed, the automatic steering control device controls the electric steering device 7 so that the steering angle follows the target steering angle. When manual driving is performed, the automatic steering control device controls the electric steering device 7 so as to assist the driver's steering operation. The light control device controls the turn signal according to an operation command for the turn signal.

[0041] <Virtual vehicle controller before restriction processing> When the control constant setting unit 52 described later performs a control process to limit the control constant within a range of limit values, the vehicle control unit 53 uses a virtual vehicle controller, which is a vehicle controller having the control constant before the limit process, to calculate the target value for virtual vehicle control.

[0042] The virtual vehicle controller is the same as the vehicle controller actually used for control, but the control constants set therein differ between the control constants before and after the limiting process, and therefore a description thereof will be omitted.

[0043] When the deviation between the actual vehicle control target value and the virtual vehicle control target value exceeds a threshold, the vehicle control unit 53 notifies the occupant of the vehicle that the change in vehicle control behavior has become large due to the limitation process. The driver is notified by a speaker or a display device of the human interface device 37. Furthermore, even when the control constant is limited, the vehicle control unit 53 may notify the occupant of the vehicle that the control constant is limited.

[0044] According to this configuration, if the control processing causes the vehicle control behavior of automatic driving or driving assistance to change significantly from the driver's vehicle control behavior, the driver will be notified, allowing the occupants to understand the situation and reducing their anxiety and irritation.

[0045] 1-1-3. Control constant setting unit 52 The control constant setting unit 52 uses characteristic information of the target driver and driving data representing the target driver's past vehicle control behavior and the surrounding environment when driving manually as input parameters, and uses a learned constant setting model that outputs control constants that will result in vehicle control behavior that corresponds to the target driver's past vehicle control behavior and the surrounding environment when driving manually as input.The control constant setting unit 52 inputs characteristic information of the driver of the vehicle itself and driving data representing the target driver's past vehicle control behavior and the surrounding environment when driving manually into the learned constant setting model, and sets the control constants output from the learned constant setting model as control constants to be used in the vehicle controller.

[0046] According to this configuration, the learned constant setting model is used to set control constants that result in vehicle control behavior that corresponds to past driving data when the driver of the vehicle is manually driving, and automatic driving control or driving assistance control of the vehicle is performed using a vehicle controller having the set control constants. Therefore, automatic driving control or driving assistance control that corresponds to manual driving by the driver of the vehicle can be performed, and the ride comfort of the driver of the vehicle can be improved.

[0047] <Trained constant setting model> As shown in Figure 3, when learning the learned constant setting model, the target driver is set to the driver to be learned, and is set to an unspecified number of drivers for whom driver characteristic information and driving data can be obtained. For each target driver, a dataset is prepared, which includes the target driver's characteristic information, past driving data of the target driver when driving manually, and control constants corresponding to the past driving data of the target driver when driving manually. Then, the learned constant setting model is trained in advance using multiple datasets of multiple target drivers. A neural network or the like is used for the model, and known machine learning or the like is used as the learning method.

[0048] A part or all of the control constant setting unit 52 may be provided in a server (arithmetic processing device) outside the vehicle. For example, the learning function of the learned constant setting model is performed by a server outside the vehicle, and the learning result (setting constants of the learned constant setting model) is transmitted via communication to the control constant setting unit 52 provided in the host vehicle and stored in a storage device. Note that the entire control constant setting unit 52 may be provided in a server outside the vehicle. In this case, the control constant setting unit 52 may transmit, via communication, to the vehicle control unit 53 provided in the host vehicle, control constants corresponding to the driver of the host vehicle calculated based on characteristic information of the driver of the host vehicle acquired from the driving assistance device 50 and past driving data of the driver of the host vehicle during manual driving. Learning of the learned constant setting model may be performed by a server (arithmetic processing device) outside the vehicle that is separate from the control constant setting unit 52.

[0049] For example, the characteristic information of the target driver is information likely to be related to driving tendencies such as age, gender, driving history, and driving preferences (e.g., safe driving, sporty driving, etc.). For example, the past vehicle control behavior and surrounding environment of the target driver when manually driving include past time-series driving conditions (speed, acceleration, yaw rate, lateral acceleration, steering angle, steering angular velocity, lateral position relative to the lane, distance between vehicles in front and behind, etc.), road information at each point in time (lane information, road type, speed limit, lane curvature, etc.), and the conditions of surrounding vehicles (relative positions, etc.). The learning arithmetic processing device or control constant setting unit 52 collects the characteristic information and driving data of the target driver in advance. This information may be collected directly from multiple vehicles via wireless communication, or previously collected data may be used.

[0050] The control constant setting unit 52 extracts driving data for a period in which a type of vehicle control associated with each of the plurality of control constants is being performed from the driving data, extracts feature values ​​associated with the control constants from the driving data for the extracted period, and sets the control constants based on the extracted feature values. For example, statistical processing is performed on the plurality of feature values ​​extracted for each control constant, and the control constants are set based on the statistical values. For example, the control constants are set based on the average value or a value corresponding to a predetermined standard deviation. The driving data for each period extracted for each control constant is used as the driving data used for learning.

[0051] As described above, the multiple control constants are the maximum allowable acceleration, the maximum allowable deceleration, the target inter-vehicle distance for each speed, the target speed for each road type or speed limit, the target lane change distance for each speed, the target lane change trajectory for each speed, the target deceleration timing and entry speed when entering a curve for each curve curvature, etc. The types and number of control constants to be learned can be changed as appropriate according to the types and number of control constants that can be set in the vehicle controller.

[0052] The types of vehicle control associated with each control constant include acceleration associated with the maximum allowable acceleration, deceleration control associated with the maximum allowable deceleration, maintaining the distance between vehicles associated with the target value of the distance between vehicles, lane changes associated with the target value of the distance required to change lanes and the target value of the lane change trajectory, and driving on a curved road associated with the target value of the deceleration timing when entering a curve and the entry speed.

[0053] For example, a period of acceleration related to the maximum allowable acceleration is extracted from the travel data, the maximum acceleration during that period is extracted, and statistical processing is performed on the extracted multiple maximum accelerations to set the maximum allowable acceleration.A period of deceleration related to the maximum allowable deceleration is extracted from the travel data, the maximum deceleration during that period is extracted, and statistical processing is performed on the extracted multiple maximum decelerations to set the maximum allowable deceleration.

[0054] A period during which an inter-vehicle distance related to a target inter-vehicle distance value is maintained is extracted from the driving data, the inter-vehicle distance and speed during that period are extracted, and statistical processing is performed on the extracted multiple inter-vehicle distances and speeds to set a target inter-vehicle distance value for each speed. A period during which lane changes are being made related to a target required lane-change distance and a target lane-change trajectory value is extracted from the driving data, and the required lane-change distance, lane-change trajectory, and speed during that period are extracted, and statistical processing is performed on the extracted multiple required lane-change distances, lane-change trajectories, and speeds to set a target required lane-change distance value for each speed and a target lane-change trajectory value for each speed.

[0055] From the driving data, a period during which the vehicle is traveling on a curved road related to the target values ​​of the deceleration timing and approach speed when entering the curve is extracted, the deceleration timing, approach speed, and curve curvature for the start point of the curve during that period are extracted, and statistical processing is performed on the extracted multiple deceleration timings, approach speeds, and curve curvatures to set the target values ​​of the deceleration timing and approach speed for each curve curvature.

[0056] The driving data used for learning is the driving data for the period extracted for calculating each control constant.

[0057] <Control constant settings> The control constant setting unit 52 acquires characteristic information of the driver of the vehicle. The characteristic information is information that the driver has registered in advance through user registration, etc. The control constant setting unit 52 stores the vehicle control behavior and the surrounding environment when the driver of the vehicle is manually driving, and uses this as driving data.

[0058] As shown in Figure 4, the control constant setting unit 52 inputs characteristic information of the driver of the vehicle and driving data representing the past vehicle control behavior and surrounding environment when the driver of the vehicle is manually driving into a learned constant setting model, and sets the control constants output from the learned constant setting model as control constants to be used in the vehicle controller.

[0059] The control constant setting unit 52 extracts driving data for a period when a type of vehicle control associated with each of the plurality of control constants is being performed from past driving data of the driver of the vehicle when the driver is driving manually, and inputs the driving data for each period extracted for each control constant into the learned constant setting model. The extraction of driving data is similar to the extraction of driving data for learning, so a description thereof will be omitted.

[0060] In this embodiment, as described above, the multiple control constants include the maximum allowable acceleration, the maximum allowable deceleration, the target inter-vehicle distance for each speed, the target speed for each road type or speed limit, the target lane change distance for each speed, the target lane change trajectory for each speed, the deceleration timing when entering a curve and the target entry speed for each curve curvature, etc.

[0061] <Control constant limit> If the control constant calculated using the learned constant setting model exceeds the limit value, the control constant setting unit 52 performs a limiting process to limit the control constant within the range of the limit value.

[0062] If the control constants are set to be the same as when the driver of the host vehicle is manually driving, the safety or driving performance of the automated driving or driving assistance may be impaired. By limiting the control constants to within the range of the limit values, the safety or driving performance of the automated driving or driving assistance can be ensured while approximating the vehicle control behavior when the driver of the host vehicle is manually driving.

[0063] The limit value of each control constant is set in advance taking into consideration the safety or driving performance of the automatic driving or driving assistance. An upper limit value and a lower limit value are set as the limit value of each control constant, and each control constant is limited to a range from the upper limit value to the lower limit value. Either the upper limit value or the lower limit value may be set.

[0064] For example, for the maximum allowable acceleration and maximum allowable deceleration, the upper limit value is set to prevent dangerous acceleration or deceleration, and the lower limit value is set to prevent acceleration or deceleration from becoming too slow.For the target value of the inter-vehicle distance for each speed, the upper limit value is set for each speed so that the inter-vehicle distance does not become too large and interfere with traffic, and the lower limit value is set for each speed so that a collision can be prevented when the preceding or following vehicle or the vehicle suddenly decelerates.For the target value of the speed for each road type or speed limit, the upper limit value is set for each road type or speed limit so that the speed does not exceed the speed limit or become a dangerous speed, and the lower limit value is set for each road type or speed limit so that the speed does not become too slow and interfere with traffic.

[0065] Regarding the target value of the required lane change distance for each speed, the upper limit is set for each speed so that the required distance is not too long and causes a traffic obstruction, and the lower limit is set for each speed so that the required distance is not too short and causes a dangerous lane change. Regarding the target value of the lane change trajectory for each speed, the upper and lower limit values ​​for the lateral movement amount relative to the travel distance after the lane change is initiated are set for each speed, taking safety and ride comfort into consideration. Regarding the deceleration timing when entering a curve for each curve curvature, the upper limit is set for each curve curvature so that the deceleration start point is not too early and causes a traffic obstruction, and the lower limit is set for each curve curvature so that the deceleration start point is not too close and causes a danger. Regarding the target value of the entry speed for each curve curvature, the upper limit is set for each curve curvature so that the speed is not too high for the curvature and causes a danger, and the lower limit is set for each curve curvature so that the speed is not too low and causes a traffic obstruction.

[0066] <Adjustment of control constants based on psychological state> The occupant state estimation unit 54 estimates the psychological state of an occupant (e.g., a driver) related to ride comfort when the host vehicle is controlled using a vehicle controller having control constants set using the learned constant setting model. The occupant state estimation unit 54 estimates the psychological state based on the output of the occupant monitoring device 38.

[0067] The occupant state estimation unit 54 evaluates the level of anxiety or fear and the level of irritation or dissatisfaction as the occupant's psychological state related to the ride comfort. The occupant state estimation unit 54 estimates the occupant's psychological state by performing known image recognition processing on an image of the occupant's face, etc., captured by a camera. The occupant state estimation unit 54 also estimates the occupant's psychological state by performing intelligent psychological analysis processing on the occupant's biometric information (heart rate, blood pressure, respiratory rate, etc.) acquired by a biometric information sensor.

[0068] If the level of anxiety or fear, or the level of irritation or dissatisfaction is equal to or greater than a judgment value, the control constant setting unit 52 stores the level of anxiety or fear, and the level of irritation or dissatisfaction in a memory device along with the vehicle control behavior and surrounding environment for the corresponding period.

[0069] The control constant setting unit 52 corrects the control constants output from the learned constant setting model based on the psychological state detected in the past.

[0070] According to this configuration, the control constants can be corrected taking into account the psychological state of the occupants when the vehicle is actually controlled using a vehicle controller having the control constants, and the corrections can be reflected in the next driving session, thereby improving ride comfort.

[0071] For example, the control constant setting unit 52 determines the type of vehicle control and control constant related to the vehicle control behavior and the surrounding environment when the level of anxiety, etc. or the level of irritation, etc. is equal to or greater than a determination value, and corrects the determined control constant output from the learned constant setting model in a direction that reduces the anxiety or irritation that is equal to or greater than the determination value. The control constant setting unit 52 may correct the control constant based on the level of the psychological state after statistical processing (e.g., averaging) of the levels of the psychological state over multiple periods obtained for each control constant. Note that if the corrected control constant is limited by the above-mentioned limiting process, the control constant does not need to be corrected.

[0072] The correction amount of each control constant is cumulatively changed from the previous correction. Furthermore, if the control constants themselves output from the learned constant setting model change due to a model update or a change in past driving data input by the driver of the vehicle during manual driving, the correction amount for the control constants output from the learned constant setting model may be changed so as to reflect the cumulative change, or the correction amount may be reset.

[0073] For example, if the level of anxiety or the like is equal to or greater than a threshold value during a period related to the maximum allowable acceleration or maximum allowable deceleration, it can be assumed that the acceleration or deceleration was too large, causing anxiety, and the maximum allowable acceleration or maximum allowable deceleration is corrected downward.On the other hand, if the level of irritation or the like is equal to or greater than a threshold value, it can be assumed that the acceleration or deceleration was too small, causing irritation, and the maximum allowable acceleration or maximum allowable deceleration is corrected upward.

[0074] If the level of anxiety or the like is equal to or greater than the judgment value during a period related to the target value of the inter-vehicle distance, it can be assumed that the inter-vehicle distance is too short and therefore the driver feels anxious, and the target value of the inter-vehicle distance is corrected to be increased. On the other hand, if the level of irritation or the like is equal to or greater than the judgment value, it can be assumed that the inter-vehicle distance is too long and therefore the driver feels irritated, and the target value of the inter-vehicle distance is corrected to be decreased.

[0075] If the level of anxiety or the like is equal to or greater than the judgment value during the period related to the target speed value, it can be assumed that the speed was too high and caused anxiety, so the target speed value is corrected downward.On the other hand, if the level of irritation or the like is equal to or greater than the judgment value, it can be assumed that the speed was too low and caused irritation, so the target speed value is corrected upward.

[0076] If the level of anxiety, etc., is equal to or greater than a judgment value during a period related to the target value of the required lane-change distance and the target value of the lane-change trajectory, it can be assumed that the lane change was too abrupt and the driver felt anxious, so the target value of the required lane-change distance is corrected upward and the change in the target value of the lane-change trajectory is corrected to be more gradual.On the other hand, if the level of irritation, etc., is equal to or greater than a judgment value, it can be assumed that the lane change was too slow and the driver felt irritated, so the target value of the required lane-change distance is corrected downward and the change in the target value of the lane-change trajectory is corrected to be more steep.

[0077] If the level of anxiety, etc. is equal to or greater than the judgment value during a period related to the target values ​​of the deceleration timing and approach speed when entering the curve, it can be assumed that the driver felt anxiety because the vehicle was driving too roughly when entering the curve, so the deceleration timing is corrected to be earlier and the approach speed is corrected to be lower.On the other hand, if the level of irritation, etc. is equal to or greater than the judgment value, it can be assumed that the driver felt irritation because the vehicle was driving too slowly when entering the curve, so the deceleration timing is corrected to be later and the approach speed is corrected to be higher.

[0078] When the vehicle is controlled again using the vehicle controller having the corrected control constants and the psychological state is obtained again, the control constants are corrected again using the obtained psychological state. As described above, it is preferable that the correction amount of each control constant is changed cumulatively.

[0079] Alternatively, the control constant setting unit 52 may re-learn the learned constant setting model based on a psychological state detected in the past, and change the control constants output from the learned constant setting model.

[0080] According to this configuration, the trained constant setting model is retrained, so that the psychological state evaluation results can be accumulated in the model. For example, the trained constant setting model is retrained so that the control constants output from the trained constant setting model approach the control constants corrected in consideration of the psychological state as described above. Then, the control constants output from the trained constant setting model after retraining are used by the vehicle controller during the next driving session.

[0081] Although exemplary embodiments are described in the present disclosure, the various features, aspects, and functions described in the embodiments are not limited to the application of a particular embodiment, but can be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are anticipated within the scope of the technology disclosed in the specification of the present disclosure. For example, this includes cases where at least one component is modified, added, or omitted. [Explanation of symbols]

[0082] 38: Occupant monitoring device, 50: Driving assistance device, driving assistance system, 51: Information acquisition unit, 52: Control constant setting unit, 53: Vehicle control unit, 54: Occupant state estimation unit

Claims

1. an information acquisition unit that acquires the running state of the host vehicle and the surrounding environment of the host vehicle; a vehicle control unit that uses a vehicle controller that performs automatic driving or driving assistance and that is capable of changing a control constant that affects a control behavior of the host vehicle based on the running state of the host vehicle and the surrounding environment, calculates a target value for vehicle control of the host vehicle, and controls the host vehicle based on the target value; a control constant setting unit that uses characteristic information of a target driver and driving data representing the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, and outputs control constants that will result in vehicle control behavior that corresponds to the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, inputs the characteristic information of the driver of the vehicle and the driving data representing the target driver's past vehicle control behavior and the surrounding environment when manually driving as input parameters, and sets the control constants output from the learned constant setting model as the control constants used in the vehicle controller; A driving assistance system equipped with

2. an occupant state estimation unit that estimates a psychological state of an occupant related to ride comfort when the host vehicle is controlled using the vehicle controller having the control constants set using the learned constant setting model; The driving assistance system of claim 1, wherein the control constant setting unit corrects the control constant output from the learned constant setting model based on the psychological state detected in the past, or re-learns the learned constant setting model to change the control constant output from the learned constant setting model.

3. The driving assistance system according to claim 2 , wherein the occupant state estimation unit estimates the psychological state based on an output of an occupant monitoring device that monitors the state of the occupant.

4. the control constant setting unit performs a limiting process to limit the control constant to a range of the limiting value when the control constant calculated using the learned constant setting model exceeds a limiting value; The driving assistance system according to claim 1 , wherein the vehicle control unit calculates a target value for the vehicle control using the vehicle controller having the control constant after the limiting process.

5. When the limiting process is performed, the vehicle control unit calculates a target value for virtual vehicle control using a virtual vehicle controller that is the vehicle controller having the control constant before the limiting process, and 5. The driving assistance system according to claim 4, wherein when a deviation between the target value of the vehicle control and the target value of the virtual vehicle control exceeds a threshold value, an occupant of the vehicle is notified that a change in the vehicle control behavior has become large due to the limiting process.

Images