Vehicle behavior estimation device and vehicle behavior estimation method

Abstract

Provided are a vehicle behavior estimation device and a vehicle behavior estimation method that are capable of, when estimating the vehicle behavior of a second vehicle present in the vicinity of a first vehicle, obtaining information about the reliability of the estimation based on the degree of difficulty of the estimation. This vehicle behavior estimation device: acquires information about the road on which the first vehicle is traveling and information about the second vehicle; estimates the vehicle behavior of the second vehicle on the basis of at least the information about the second vehicle and the road information; and calculates the reliability of the estimation of the vehicle behavior of the second vehicle using the degree of difficulty of the estimation of the vehicle behavior of the second vehicle on the basis of at least one of information about the first vehicle, the information about the second vehicle, and the road information.

Description

Vehicle behavior estimation device and vehicle behavior estimation method

[0001] The present disclosure relates to a vehicle behavior estimation device and a vehicle behavior estimation method.

[0002] There is a technology that improves safety and comfort by predicting a lane change by another vehicle that cuts in front of the driver's vehicle before the cut-in is completed and controlling the driver's vehicle in advance. In the technology disclosed in Patent Document 1, the lane change of the other vehicle is predicted from the detected driving state of the other vehicle, such as the position and speed of the other vehicle. In addition, the judgment criteria are changed depending on the distance between the driver's vehicle and the other vehicle.

[0003] Japanese Patent Application Laid-Open No. 2021-142901

[0004] However, the difficulty of estimating the vehicle behavior of the other vehicle changes depending on the detection state of the other vehicle's traveling state and the road conditions. For example, when the detection state of the sensor that detects the traveling state of the other vehicle is poor or when the visibility of the other vehicle from the own vehicle is poor, the difficulty of estimation increases and the reliability of the estimation result decreases.

[0005] However, in the technology of Patent Document 1, lane changes are estimated from the detected driving states of other vehicles, but the difficulty of the estimation is not taken into consideration. Therefore, even when the difficulty of the estimation is high and the reliability of the estimation result is low, the estimation result is still used.

[0006] Therefore, an object of the present disclosure is to provide a vehicle behavior estimation device and a vehicle behavior estimation method that can obtain information on the estimation reliability based on the difficulty of the estimation when estimating the vehicle behavior of a second vehicle that is present around a first vehicle.

[0007] The vehicle behavior estimation device according to the present disclosure includes an information acquisition unit that acquires road information on which a first vehicle is traveling and information on surrounding vehicles present around the first vehicle; a second vehicle behavior estimation unit that sets a second vehicle, the vehicle behavior of which is to be estimated, from the surrounding vehicles and estimates the vehicle behavior of the second vehicle based on at least the information on the second vehicle and the road information; and a reliability calculation unit that calculates the estimation reliability of the vehicle behavior of the second vehicle according to the degree of difficulty in estimating the vehicle behavior of the second vehicle, based on at least one of the information on the first vehicle, the information on the second vehicle, and the road information.

[0008] The vehicle behavior estimation method according to the present disclosure includes an information acquisition step of acquiring road information on which a first vehicle is traveling and information on surrounding vehicles present around the first vehicle; a second vehicle behavior estimation step of setting a second vehicle, the vehicle behavior of which is to be estimated, from the surrounding vehicles and estimating the vehicle behavior of the second vehicle based on at least the information on the second vehicle and the road information; and a reliability calculation step of calculating the estimation reliability of the vehicle behavior of the second vehicle according to the degree of difficulty in estimating the vehicle behavior of the second vehicle, based on at least one of the information on the first vehicle, the information on the second vehicle, and the road information.

[0009] According to the vehicle behavior estimation device and vehicle behavior estimation method of the present disclosure, the estimation reliability of the vehicle behavior of the second vehicle is calculated based on the difficulty of estimating the vehicle behavior of the second vehicle, based on at least one of the information of the first vehicle, the information of the second vehicle, and the road information, so that an evaluation index can be given to the estimation result of the vehicle behavior of the second vehicle, and the estimation result of the vehicle behavior of the second vehicle can be used appropriately.

[0010] 1 is a schematic block diagram of a vehicle behavior estimation device according to a first embodiment. FIG. 2 is a schematic hardware configuration diagram of a vehicle behavior estimation device according to a first embodiment. FIG. 3 is a schematic hardware configuration diagram of a vehicle behavior estimation device according to a first embodiment. FIG. 4 is a diagram for explaining a coordinate system of a first vehicle and a road coordinate system of a second vehicle according to the first embodiment. FIG. 5 is a schematic diagram for explaining a lane change by a second vehicle to the front of the first vehicle according to the first embodiment. FIG. 6 is a schematic diagram for explaining calculation of estimation reliability of each factor according to the first embodiment. FIG. 7 is a flowchart for explaining processing by the vehicle behavior estimation device according to the first embodiment. FIG. 8 is a schematic block diagram of a vehicle behavior estimation device according to a second embodiment. FIG. 9 is a diagram for explaining determination by a vehicle behavior determination unit according to the second embodiment. FIG. 10 is a diagram for explaining determination by a vehicle behavior determination unit according to the second embodiment. FIG. 11 is a flowchart for explaining processing by the vehicle behavior estimation device according to the second embodiment.

[0011] 1. First Embodiment A vehicle behavior estimation device 1 according to a first embodiment will be described with reference to the drawings. In this embodiment, the vehicle behavior estimation device 1 is provided in a first vehicle.

[0012] As shown in FIG. 1, the first 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 vehicle behavior estimation device 1, a drive control device 36, a power motor 8, an electric steering device 7, and an electric braking device 9, etc.

[0013] The periphery monitoring device 31 is a device such as a camera or radar that monitors the periphery of the first vehicle (at least the area ahead of the first vehicle). The radar may be a millimeter wave radar, a laser radar, an ultrasonic radar, etc. The wireless communication device 35 performs wireless communication with base stations, surrounding vehicles, roadside devices, etc. using a cellular wireless communication standard such as 4G or 5G.

[0014] The position detection device 32 is a device that detects the current position (latitude, longitude, altitude) of the first vehicle, and uses a GPS antenna or the like that receives signals output from artificial satellites such as the Global Navigation Satellite System (GNSS). Note that various methods may be used to detect the current position of the first vehicle, such as a method using the lane number of the first vehicle, a map matching method, a dead reckoning method, or a method using detected information around the first vehicle.

[0015] The map information database 34 stores road information such as road shape (e.g., number of lanes, position of each lane, shape of each lane, type of each lane, road type, speed limit, etc.), signs, traffic lights, etc. The shape of the lane includes the position, width, curvature, gradient (longitudinal gradient, lateral gradient), etc. of each point along the direction of travel. The map information database 34 is mainly composed of a storage device. Note that the map information database 34 may be provided in a server outside the vehicle connected to a network, and the vehicle behavior estimation device 1 may obtain necessary road information from the server outside the vehicle via the wireless communication device 35.

[0016] The drive control device 36 includes a power control device, a brake 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 brake control device controls the braking operation of the electric brake device 9. The automatic steering control device controls the electric steering device 7. The light control device controls turn signals, hazard lights, etc.

[0017] The vehicle state detection device 33 is a detection device that detects a first vehicle state, which is the driving state and traveling state of the first vehicle. In the present embodiment, the vehicle state detection device 33 detects the speed, acceleration, yaw rate, steering angle, lateral acceleration, etc. of the first vehicle as the traveling state of the first vehicle. For example, the vehicle state detection device 33 may be provided with a speed sensor, an acceleration sensor, an angular velocity sensor, a steering angle sensor, etc. that detect the rotational speed of the wheels.

[0018] The vehicle state detection device 33 detects the acceleration / deceleration operation, steering angle operation, and lane change operation by the driver as the driving state of the first vehicle. For example, an accelerator position sensor, a brake position sensor, a steering angle sensor (steering wheel angle sensor), a steering torque sensor, a turn signal position switch, etc.

[0019] 1-1. Vehicle behavior estimation device 1 The vehicle behavior estimation device 1 includes functional units such as an information acquisition unit 51, a second vehicle behavior estimation unit 52, and a reliability calculation unit 53. Each function of the vehicle behavior estimation device 1 is realized by a processing circuit included in the vehicle behavior estimation device 1. Specifically, as shown in Fig. 2, the vehicle behavior estimation device 1 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.

[0020] 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. As the storage device 91, various 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. may be used.

[0021] The input / output device 92 includes 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, etc., and communicates with these devices.

[0022] The functions of the functional units 51 to 53 included in the vehicle behavior estimation device 1 are realized by the arithmetic processing device 90 executing software (programs) stored in the storage device 91 and cooperating with other hardware of the vehicle behavior estimation device 1, such as the storage device 91 and the input / output device 92. Note that setting data used by the functional units 51 to 53 is stored in the storage device 91, such as an EEPROM.

[0023] 3, the vehicle behavior estimation device 1 may be provided with dedicated hardware 93 as a processing circuit, such as a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC, an FPGA, a GPU, an AI chip, or a circuit combining these. Each function of the vehicle behavior estimation device 1 will be described in detail below.

[0024] 1-1-1. Information Acquisition Unit 51 The information acquisition unit 51 acquires road information on which the first vehicle is traveling. In this embodiment, the information acquisition unit 51 acquires road information around the first vehicle from the map information database 34 based on the position information of the first 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. The shape of the lane includes the position, width, curvature, gradient (longitudinal gradient, lateral gradient), etc. of each point along the direction of travel.

[0025] In addition, the information acquisition unit 51 detects the shape and type of road dividing lines, etc. based on the detection information of white lines, road shoulders, etc. acquired from the perimeter monitoring device 31, and determines the shape, number of lanes, type of each lane, etc. based on the detected shape and type of road dividing lines, etc.

[0026] The information acquisition unit 51 acquires lane information corresponding to the lane in which the first vehicle is traveling, based on the position of the first vehicle. The information acquisition unit 51 also acquires lane information corresponding to the lane in which each of the surrounding vehicles is traveling, based on the positions of each of the surrounding vehicles. The acquired lane information includes the shape, position, and type of the lane, as well as lane information of the surrounding lanes.

[0027] The information acquisition unit 51 acquires the status of surrounding vehicles present around the first vehicle. In this embodiment, the information acquisition unit 51 acquires the position, movement direction, speed, acceleration, etc. of the surrounding vehicles based on the detection information acquired from the periphery monitoring device 31 and the position information of the first vehicle acquired from the position detection device 32. In addition to the surrounding vehicles, the information acquisition unit 51 also acquires information on obstacles, pedestrians, signs, traffic regulations such as lane restrictions, etc.

[0028] In this embodiment, the information acquisition unit 51 acquires the relative positions and relative speeds of surrounding vehicles and the like with respect to the first vehicle in a coordinate system of the first vehicle based on the current position of the first vehicle. As shown in Fig. 4, the coordinate system of the first vehicle has axes in the longitudinal direction X of the current first vehicle and the lateral direction Y of the first vehicle.

[0029] 1-1-2. Second vehicle behavior estimation unit 52 The second vehicle behavior estimation unit 52 sets a second vehicle, whose vehicle behavior is to be estimated, from among the surrounding vehicles, and estimates the vehicle behavior of the second vehicle based on at least information about the second vehicle and information about the road on which the first vehicle is traveling.

[0030] The second vehicle behavior estimation unit 52 sequentially sets, among the surrounding vehicles, vehicles that affect the traveling of the first vehicle as the second vehicle, and estimates the vehicle behavior. For example, each surrounding vehicle traveling within a determination distance range from the first vehicle in the traveling lane of the first vehicle and an adjacent lane is set as the second vehicle. For each surrounding vehicle set as the second vehicle, the vehicle behavior is estimated, and the estimation reliability of the vehicle behavior is calculated. Note that the second vehicle may be limited to the surrounding vehicles traveling in each lane ahead of the first vehicle.

[0031] In this embodiment, the second vehicle behavior estimator 52 estimates the probability of the second vehicle changing lanes as the vehicle behavior of the second vehicle. The probability of changing lanes to the left lane and the probability of changing lanes to the right lane are estimated separately. For example, the probability of changing lanes to the left lane, the probability of changing lanes to the right lane, and the probability of maintaining the lane are estimated so that the sum of these is 100%. In particular, as shown in FIG. 5 , the second vehicle behavior estimator 52 may be configured to estimate the probability of the second vehicle changing lanes to the driving lane of the first vehicle, ahead of the first vehicle.

[0032] <Probability of lane change due to lateral movement of second vehicle> For example, the second vehicle behavior estimation unit 52 calculates information regarding the lateral movement of the second vehicle based on information about the second vehicle and road information, and estimates the probability of lane change due to lateral movement of the second vehicle based on the information regarding the lateral movement.

[0033] The second vehicle behavior estimation unit 52 calculates at least one of the following information regarding the lateral movement of the second vehicle relative to the lane in which the second vehicle is traveling: the lateral position, speed, acceleration, the change in position over time, the change in speed over time, and the change in acceleration over time of the second vehicle relative to the lane in which the second vehicle is traveling, and based on the calculated information regarding the lateral movement, the greater the tendency of the second vehicle to move laterally relative to the lane in which the second vehicle is traveling, the higher the probability of a lane change due to the estimated lateral movement.

[0034] According to this configuration, the probability of a lane change due to lateral movement can be accurately estimated based on the information regarding the lateral movement of the second vehicle as described above.

[0035] 4 , in this embodiment, the second vehicle behavior estimator 52 converts the relative position, relative speed, and relative acceleration of the second vehicle in the coordinate system of the first vehicle into the relative position, relative speed, and relative acceleration of the second vehicle in a coordinate system of the longitudinal direction S and lateral direction W of the lane on which the second vehicle is located (referred to as the road coordinate system of the second vehicle). The relative position before and after the conversion includes the longitudinal relative position and the lateral relative position in each coordinate system, the relative speed before and after the conversion includes the longitudinal relative speed and the lateral relative speed in each coordinate system, and the relative acceleration before and after the conversion includes the longitudinal relative acceleration and the lateral relative acceleration in each coordinate system.

[0036] In the road coordinate system of the second vehicle, the tangent direction to the center line of the lane in which the second vehicle is traveling is taken as the longitudinal direction S, and the normal direction to the center line is taken as the lateral direction W. The position on the center line that is closest to the position of the second vehicle is taken as the origin.

[0037] The second vehicle behavior estimation unit 52 performs a rotational coordinate transformation on the relative position, relative speed, and relative acceleration of the second vehicle in the coordinate system of the first vehicle based on the relative angle of the longitudinal direction S of the road coordinate system of the second vehicle with respect to the longitudinal direction X of the coordinate system of the first vehicle, and a translation transformation based on the relative position of the origin of the road coordinate system of the second vehicle with respect to the origin of the coordinate system of the first vehicle, to convert the relative position, relative speed, and relative acceleration of the second vehicle in the road coordinate system of the second vehicle. These rotational coordinate transformations and translation transformations are well-known techniques, and therefore will not be described here.

[0038] The second vehicle behavior estimation unit 52 calculates time difference values ​​of the converted relative position, relative velocity, and relative acceleration in the lateral direction W, thereby calculating the amount of change over time of each of the converted relative position, relative velocity, and relative acceleration in the lateral direction W. For example, as the time difference values ​​(amount of change over time), a difference value between a current value and a value from a predetermined time ago or a difference value between a current value and a filtered value (for example, a moving average value) is calculated.

[0039] <Use of learning model> The second vehicle behavior estimation unit 52 uses a learning model that is trained to receive information regarding the lateral movement of the second vehicle and output the probability of a lane change due to lateral movement, and determines the probability of a lane change due to lateral movement of the second vehicle based on the information regarding the lateral movement of the second vehicle.

[0040] For example, a logistic regression model is used as the learning model. The parameters of the logistic regression model are adjusted in advance using a data set of information on the lateral movement of the second vehicle and whether or not the vehicle has changed lanes, which has been collected in advance, so that the information on the lateral movement of the second vehicle can be separated into whether or not the vehicle has changed lanes. When a lane change has occurred, the probability of the lane change is 100%, and when a lane change has not occurred, the probability of the lane change is 0%.

[0041] Alternatively, various known learning models such as a neural network may be used as the learning model. Furthermore, the probability of a lane change may be estimated instead of the presence or absence of a lane change (100% or 0%).

[0042] The second vehicle behavior estimation unit 52 may estimate the probability of the second vehicle changing lanes based on the results of comparing each piece of information related to the lateral movement of the second vehicle with a threshold value set for each piece of information.

[0043] In addition, the second vehicle behavior estimation unit 52 estimates the relative position of the second vehicle to the first vehicle after the lane change, i.e., the relative position of the lane change destination, based on information regarding the lateral movement of the second vehicle and the relative position and relative speed between the first vehicle and the second vehicle.

[0044] <Probability Modification Based on Surrounding Environment of Second Vehicle> The second vehicle behavior estimation unit 52 may modify the probability of a lane change due to lateral movement of the second vehicle based on the relationship between the second vehicle and the surrounding environment of the second vehicle. For example, if there is a condition on the left side of the second vehicle that would be an obstacle to a lane change (for example, if there is no adjacent lane on the left or if an obstacle to a lane change exists in the adjacent lane on the left), the second vehicle behavior estimation unit 52 modifies the probability of a lane change to the left lane to 0% and increases the probability of a lane change to the right lane and the probability of maintaining the lane so that the sum of these probabilities becomes 100%. The same applies to the right lane.

[0045] <Probability of lane change due to triggering factor> The second vehicle behavior estimation unit 52 may estimate the probability of a lane change due to a triggering factor that triggers a lane change by the second vehicle based on the relationship between the second vehicle and the surrounding environment of the second vehicle, and calculate the probability of a lane change by the second vehicle as the vehicle behavior of the second vehicle by taking a weighted average of the probability of a lane change due to lateral movement and the probability of a lane change due to a triggering factor.

[0046] One factor that may induce a lane change by the second vehicle is merging from a merging lane to a main lane. The probability of a lane change due to merging is determined in a different way from a lane change due to normal lateral movement. For example, the second vehicle behavior estimation unit 52 predicts whether the second vehicle will merge around (e.g., in front of or behind) the first vehicle based on at least the relative position and relative speed between the second vehicle traveling in the merging lane and the first vehicle traveling on the main lane, and the shape of the merging lane (such as the positions of the start and end points of the merging lane). If the second vehicle will merge, it predicts the relative merging position of the second vehicle with respect to the first vehicle. An example of a case in which the second vehicle will not merge around the first vehicle is when the second vehicle slows down or stops temporarily near the end point of the merging lane. Various known methods are used for merging prediction.

[0047] If the second vehicle behavior estimation unit 52 predicts that the second vehicle will merge around the first vehicle (for example, within a determination distance range in front of or behind the first vehicle), it sets the probability of a lane change due to merging to 100%, and if it predicts that the second vehicle will not merge around the first vehicle, it sets the probability of a lane change due to merging to 0%.If it is possible to continuously predict the probability that the second vehicle will merge around the first vehicle, the probability of a lane change due to merging is continuously set within a range from 0% to 100%.

[0048] When the second vehicle is traveling in a merging lane and the probability of a lane change due to merging is estimated, the second vehicle behavior estimation unit 52 may calculate the probability of a lane change of the second vehicle as the vehicle behavior of the second vehicle by taking a weighted average of the probability of a lane change due to lateral movement and the probability of a lane change due to merging, as shown in the following formula: Probability of lane change = Probability of lane change due to lateral movement × α + Probability of lane change due to merging × (1 - α)

[0049] Here, α is a weight for the weighted average, and is set to an appropriate value between 0 and 1. If the second vehicle is traveling in a merging lane and the probability of a lane change due to lateral movement cannot be estimated, α is set to 0, and if the second vehicle is not traveling in a merging lane and the probability of a lane change due to merging cannot be estimated, α is set to 1.

[0050] The reliability calculation unit 53 calculates the reliability of the estimation of the vehicle behavior of the second vehicle according to the degree of difficulty of estimating the vehicle behavior of the second vehicle, based on at least one of the information on the first vehicle, the information on the second vehicle, and the road information.

[0051] For example, the more difficult it is to estimate the vehicle behavior of the second vehicle, the lower the estimation reliability. An evaluation index can be provided for the estimation result of the vehicle behavior of the second vehicle by the second vehicle behavior estimator 52.

[0052] The reliability calculation unit 53 calculates the estimation reliability based on at least one of the factors contributing to the estimation difficulty: the distance between the first vehicle and the second vehicle, the road curvature related to the detection of the second vehicle, the road gradient related to the detection of the second vehicle, and the degree of overlap between the second vehicle and other vehicles related to the detection of the second vehicle.

[0053] As the distance between the first vehicle and the second vehicle increases, the accuracy of detection of the second vehicle by the periphery monitoring device 31 decreases, increasing the difficulty of estimation and lowering the reliability of estimation. As the road curvature related to the detection of the second vehicle increases, visibility of the road in front of or behind the first vehicle decreases, or the lateral fluctuation of the second vehicle increases, making it more difficult for the periphery monitoring device 31 to detect the second vehicle, increasing the difficulty of estimation and lowering the reliability of estimation. It is sufficient to use the road curvature corresponding to the road portion between the first vehicle and the second vehicle or the road portion of the second vehicle. For example, the average value of the road curvature of the road portion between the first vehicle and the second vehicle, or the road curvature of the road portion of the second vehicle, may be used.

[0054] As the road gradient related to the detection of the second vehicle increases, the visibility of the road in front of or behind the first vehicle becomes poorer, or as the lateral gradient increases, the lateral fluctuation of the second vehicle increases, making it more difficult for the periphery monitoring device 31 to detect the second vehicle, thereby increasing the difficulty of estimation and reducing the reliability of estimation. It is sufficient to use the road gradient corresponding to the road portion between the first vehicle and the second vehicle, and for example, the average value of the road gradient of the road portion between the first vehicle and the second vehicle, or the road gradient of the road portion of the second vehicle, may be used.

[0055] As the degree of overlap between the second vehicle and other vehicles involved in the detection of the second vehicle increases, the portion of the second vehicle detected by the surrounding monitoring device 31 decreases, making it more difficult to detect the second vehicle accurately, thereby increasing the difficulty of estimation and reducing the reliability of estimation.

[0056] In the present embodiment, reliability calculation unit 53 calculates the estimation reliability based on the detection states of the sensors that detect the second vehicle as factors that contribute to the estimation difficulty. The detection states of the sensors include one or more of reliability information on the detection results output from each sensor that detects the second vehicle, the number of sensors that can use the detection results among the multiple sensors that detect the second vehicle, the presence or absence of outliers in the detection values ​​of the multiple sensors that detect the second vehicle, and the presence or absence of temporal discontinuities in the detection values ​​of the multiple sensors that detect the second vehicle.

[0057] The sensors used to detect the second vehicle include a camera and radar (millimeter-wave radar, laser radar, ultrasonic radar, etc.) of the perimeter monitoring device 31. Regarding reliability information of the detection result, in the case of a camera, the reliability information of the detection result from an image changes depending on the imaging conditions (e.g., imaging conditions due to brightness, backlight, weather (fog, rain, snow), etc.). In the case of a radar, the reliability information of the detection result from radar reflection changes depending on the conditions of the subject (e.g., the reflectivity of the detected object, conditions due to weather (fog, rain, snow), etc.).

[0058] As the number of sensors whose detection results can be used decreases, the detection accuracy of the second vehicle decreases, increasing the difficulty of estimation and lowering the reliability of estimation. For example, if a sensor malfunctions or the detection status of a sensor deteriorates due to some factor, the detection results of a specific sensor may become unusable.

[0059] If, among the information on the second vehicle detected by each sensor, there is detection information from a sensor that deviates from the detection information of other sensors by more than a predetermined value, there is an outlier, which increases the difficulty of estimation and reduces the reliability of estimation.

[0060] If the information about the second vehicle detected by each sensor becomes discontinuous over time, it can be determined that the detection accuracy has deteriorated due to some factor, increasing the difficulty of estimation and reducing the reliability of estimation.

[0061] The reliability calculation unit 53 reduces the estimation reliability when the second vehicle behavior estimation unit 52 erroneously estimates the past vehicle behavior.

[0062] For example, the reliability calculation unit 53 reduces the estimation reliability when the second vehicle does not change lanes even after a determination period has elapsed after the probability of a lane change of the second vehicle estimated by the second vehicle behavior estimation unit 52 becomes equal to or greater than a predetermined value. Alternatively, the reliability calculation unit 53 reduces the estimation reliability when the second vehicle changes lanes in a state in which the second vehicle behavior estimation unit 52 estimates that the probability of a lane change of the second vehicle is equal to or less than a predetermined value. The reliability calculation unit 53 may continuously reduce the estimation reliability when a state in which the estimation result of the past vehicle behavior remains erroneous, and may continuously increase the estimation reliability when a state in which the estimation result of the past vehicle behavior remains correct.

[0063] <Comprehensive Evaluation of Each Factor> In this embodiment, the reliability calculation unit 53 calculates the estimation reliability of each factor that contributes to the estimation difficulty, and calculates the final estimation reliability based on the estimation reliability of each factor. That is, the reliability calculation unit 53 comprehensively evaluates the estimation reliability of each factor to calculate the final estimation reliability.

[0064] For example, the reliability calculation unit 53 calculates the final estimation reliability Rall by multiplying the estimation reliability R of each factor as shown in the following formula: Rall = R1 × R2 × R3 × R4 × R5 × R6 Here, for example, R1 is the estimation reliability calculated based on the distance between the first vehicle and the second vehicle. R2 is the estimation reliability calculated based on the road curvature related to the detection of the second vehicle. R3 is the estimation reliability calculated based on the road gradient related to the detection of the second vehicle. R4 is the estimation reliability calculated based on the degree of overlap between the second vehicle and other vehicles. R5 is the estimation reliability calculated based on the detection state of the sensor detecting the second vehicle, and may be calculated for each type of sensor detection state or may be calculated by comprehensively evaluating multiple types of sensor detection states. R6 is the estimation reliability calculated based on errors in the estimation results of past vehicle behavior. Not all of R1 to R6 may be used, and one or more of R1 to R6 may be used.

[0065] As shown in Figure 6, as the input value U of each factor, such as distance, road curvature, road gradient, degree of overlap, degree of deterioration of the sensor detection state, and degree of error in the estimation result, increases, the estimation reliability R of each factor decreases from 1 to 0. The highest value of estimation reliability R is 1, and the lowest value of estimation reliability R is 0. The estimation reliability R is set to 1 until the input value U of each factor exceeds a threshold, and under normal conditions, the estimation reliability R of each factor is maintained at a high value and does not decrease unnecessarily. The estimation reliability R of each factor is calculated using map data or a formula preset for each factor.

[0066] Alternatively, the reliability calculation unit 53 may calculate the final estimated reliability Rall based on the estimated reliability R of each factor using a calculation method other than summation. For example, a high-order mathematical formula, a neural network, or the like may be used.

[0067] The estimated vehicle behavior of the second vehicle (in this example, the probability of the second vehicle changing lanes) and the estimated reliability of the vehicle behavior of the second vehicle are transmitted to an external device or, as in the second and third embodiments described below, to other processing units of the vehicle behavior estimation device 1, and are used for various types of processing.

[0068] <Flowchart> Fig. 7 is a schematic flowchart illustrating the processing (vehicle behavior estimation method) of the vehicle behavior estimation device 1 according to this embodiment. The processing in Fig. 7 is executed, for example, at every predetermined calculation cycle.

[0069] In step S01, as described above, the information acquisition unit 51 acquires information about the road on which the first vehicle is traveling and information about surrounding vehicles present around the first vehicle.

[0070] In step S02, as described above, the second vehicle behavior estimator 52 sets a second vehicle, whose vehicle behavior is to be estimated, from among the surrounding vehicles, and estimates the vehicle behavior of the second vehicle based on at least information about the second vehicle and information about the road on which the first vehicle is traveling. In this embodiment, the second vehicle behavior estimator 52 estimates the probability of the second vehicle changing lanes as the vehicle behavior of the second vehicle.

[0071] In step S03, as described above, the reliability calculation unit 53 calculates the estimated reliability of the vehicle behavior of the second vehicle based on the difficulty of estimating the vehicle behavior of the second vehicle, based on at least one of the information of the first vehicle, the information of the second vehicle, and the road information.

[0072] 2. Second Embodiment Next, a vehicle behavior estimation device 1 according to a second embodiment will be described. Description of the same components as those in the first embodiment will be omitted. The basic configuration of the vehicle behavior estimation device 1 according to this embodiment is the same as that of the first embodiment, but differs from the first embodiment in that the vehicle behavior estimation device 1 further includes a vehicle behavior determination unit 54, as shown in FIG. 8 .

[0073] The vehicle behavior determination unit 54 makes a final determination as to whether the second vehicle will execute the estimated vehicle behavior based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52 and the estimation reliability calculated by the reliability calculation unit 53.

[0074] According to this configuration, the final determination of whether the second vehicle will execute the estimated vehicle behavior is made taking into account the estimation reliability, thereby improving the accuracy of the final prediction of the vehicle behavior of the second vehicle.

[0075] In this embodiment, the vehicle behavior determination unit 54 makes a final determination as to whether the second vehicle will execute the estimated lane change based on the probability of the second vehicle changing lanes estimated by the second vehicle behavior estimation unit 52 as the vehicle behavior of the second vehicle and the estimation reliability calculated by the reliability calculation unit 53.

[0076] For example, as shown in Fig. 9, the vehicle behavior determination unit 54 determines that the second vehicle will make the estimated lane change if the probability of the second vehicle changing lanes exceeds a first threshold and the estimation reliability exceeds a second threshold, and otherwise determines that the second vehicle will not make the estimated lane change. In Fig. 9, a 100% probability of a lane change is represented by 1.

[0077] 10 , the vehicle behavior determination unit 54 sets a threshold value based on the estimation reliability, and when the probability of the second vehicle changing lanes exceeds the threshold value, the vehicle behavior determination unit 54 finally determines that the second vehicle will make the estimated lane change, and otherwise finally determines that the second vehicle will not make the estimated lane change. As the estimation reliability decreases, the first threshold value is increased.

[0078] Alternatively, as shown in FIG. 11, the vehicle behavior determination unit 54 may determine that the second vehicle will make the estimated lane change if the product of the probability of the second vehicle changing lanes and the estimation reliability exceeds a threshold value, and may otherwise determine that the second vehicle will not make the estimated lane change.

[0079] Alternatively, the boundary line (threshold) between the probability of the second vehicle changing lanes and the estimation reliability for determining whether the second vehicle will perform the estimated lane change may be set to any shape.

[0080] The final determination result of whether or not the estimated vehicle behavior of the second vehicle will be performed (in this example, the final determination result of whether or not the estimated second vehicle will change lanes) is transmitted to an external device or to another processing unit of the vehicle behavior estimation device 1 and is used for various processes. For example, the vehicle behavior of the second vehicle that is finally determined to be performed may be notified to the driver of the first vehicle via a human interface device such as a display device or a speaker. Furthermore, the vehicle control unit 55 described in the third embodiment may calculate the control amount of the first vehicle based on the vehicle behavior of the second vehicle that is finally determined to be performed.

[0081] <Flowchart> Fig. 12 is a schematic flowchart illustrating the processing (vehicle behavior estimation method) of the vehicle behavior estimation device 1 according to this embodiment. The processing in Fig. 12 is executed, for example, at every predetermined calculation cycle.

[0082] Steps S11 to S13 are the same as steps S01 to S03 in the first embodiment shown in FIG. 7, and therefore a description thereof will be omitted.

[0083] In step S14, as described above, the vehicle behavior determination unit 54 makes a final determination as to whether or not the second vehicle will execute the estimated vehicle behavior, based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52 and the estimation reliability calculated by the reliability calculation unit 53. In this embodiment, the vehicle behavior determination unit 54 makes a final determination as to whether or not the second vehicle will execute the estimated lane change, based on the probability of the lane change of the second vehicle estimated as the vehicle behavior of the second vehicle by the second vehicle behavior estimation unit 52 and the estimation reliability calculated by the reliability calculation unit 53.

[0084] 3. Embodiment 3 Next, a vehicle behavior estimation device 1 according to embodiment 3 will be described. Description of components that are the same as those in embodiment 1 above will be omitted. The basic configuration of the vehicle behavior estimation device 1 according to this embodiment is the same as that of embodiment 1, but differs from embodiment 1 in that the vehicle behavior estimation device 1 further includes a vehicle control unit 55, as shown in FIG. 13 .

[0085] The vehicle control unit 55 calculates the control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52 and the estimation reliability calculated by the reliability calculation unit 53.

[0086] According to this configuration, the control amount of the first vehicle is calculated based on the estimated vehicle behavior of the second vehicle, taking into account the estimation reliability, thereby improving the calculation accuracy of the control amount of the first vehicle.

[0087] In this embodiment, the vehicle control unit 55 calculates the control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52, and changes the magnitude of the calculated control amount of the first vehicle based on the estimation reliability calculated by the reliability calculation unit 53.

[0088] For example, the vehicle control unit 55 reduces the magnitude of the calculated control amount of the first vehicle as the estimation reliability decreases. The control amount may be reduced in stages or continuously in accordance with the reduction in estimation reliability. This configuration prevents the control amount of the first vehicle from increasing based on a vehicle behavior with low estimation reliability, prevents unnecessary vehicle control from being performed, and reduces discomfort felt by the driver of the first vehicle.

[0089] In this embodiment, the vehicle behavior of the second vehicle is estimated as a probability of a lane change by the second vehicle. The vehicle control unit 55 determines whether or not control of the first vehicle is necessary based on the probability of a lane change by the second vehicle. The vehicle control unit 55 determines that control of the first vehicle is necessary when the second vehicle changes lanes in front of or behind the first vehicle in the driving lane of the first vehicle and the probability of the lane change by the second vehicle exceeds a threshold value; otherwise, the vehicle control unit 55 determines that control of the first vehicle is not necessary. Alternatively, similar to the vehicle behavior determination unit 54 in the second embodiment, the vehicle control unit 55 may determine whether or not the second vehicle will execute the estimated lane change based on the probability of a lane change by the second vehicle and the estimation reliability, determine that a lane change will be executed, and determine that control of the first vehicle is necessary when the second vehicle changes lanes in front of or behind the first vehicle in the driving lane of the first vehicle, and otherwise determine that control of the first vehicle is not necessary.

[0090] When the vehicle control unit 55 determines that control of the first vehicle is necessary, it calculates the control amount of the first vehicle based on the prediction result of the lane change of the second vehicle. At this time, the vehicle control unit 55 reduces the magnitude of the calculated control amount of the first vehicle as the estimation reliability decreases. Note that, similar to the vehicle behavior determination unit 54 in the second embodiment, when it is determined whether or not to execute a lane change based on the probability of a lane change and the estimation reliability, the vehicle control unit 55 does not need to change the magnitude of the control amount of the first vehicle based on the estimation reliability.

[0091] For example, if the second vehicle is in the same lane as the first vehicle and is predicted to change lanes in front of or behind the first vehicle, the vehicle control unit 55 calculates the amount of change in the target output torque and the target braking force to increase or decrease the speed of the first vehicle based on the relative positions and relative speeds of the first and second vehicles so as to maintain a vehicle-to-vehicle distance of at least a set distance between the first and second vehicles. As the estimation reliability decreases, the vehicle control unit 55 reduces the maximum values ​​of the calculated amount of change in the target output torque and the target braking force.

[0092] The vehicle control unit 55 transmits the changed target output torque and target braking force to a power control device and a brake control device, etc., serving as the drive control device 36. The power control device then controls the output torque of the power machine 8, such as an internal combustion engine or a motor, in accordance with the target output torque. Furthermore, the brake control device controls the braking operation of the electric brake device 9 in accordance with the target braking force.

[0093] Alternatively, if the second vehicle is in the same lane as the first vehicle and is predicted to change lanes in front of or behind the first vehicle, and it is determined that the first vehicle will change lanes to avoid the second vehicle, the vehicle control unit 55 may calculate a target steering angle for changing lanes. The vehicle control unit 55 transmits the target steering angle to the automatic steering control device, and the automatic steering control device controls the electric steering device 7 so that the steering angle follows the target steering angle.

[0094] Alternatively, to perform more advanced autonomous driving, the vehicle control unit 55 may generate a time-series target driving trajectory based on the estimated vehicle behavior of the second vehicle and the estimation reliability. The time-series target driving trajectory is a time-series driving plan including the target position, target traveling direction, target speed, and target acceleration of the first vehicle at each future time. For example, when increasing the speed of the first vehicle, the vehicle control unit 55 increases the target speed and target acceleration at each future time. In this case, the increase amount is reduced as the estimation reliability decreases. The vehicle control unit 55 controls the vehicle so that the first vehicle drives along the target driving trajectory. For example, the vehicle control unit 55 determines a target output torque, a target braking force, a target steering angle, a turn signal operation command, etc., and transmits each determined command value to a power control unit, a brake control unit, an automatic steering control unit, a light control unit, etc., which serve as the drive control unit 36.

[0095] <Flowchart> Fig. 14 is a schematic flowchart illustrating the processing (vehicle behavior estimation method) of the vehicle behavior estimation device 1 according to this embodiment. The processing in Fig. 14 is executed, for example, at every predetermined calculation cycle.

[0096] Steps S21 to S23 are the same as steps S01 to S03 in the first embodiment shown in FIG. 7, and therefore a description thereof will be omitted.

[0097] In step S24, as described above, the vehicle control unit 55 calculates the control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52 and the estimation reliability calculated by the reliability calculation unit 53. In this embodiment, the vehicle control unit 55 calculates the control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit 52, and changes the magnitude of the calculated control amount of the first vehicle based on the estimation reliability calculated by the reliability calculation unit 53.

[0098] In the above-described embodiments, the second vehicle behavior estimation unit 52 estimates the probability of the second vehicle changing lanes as the vehicle behavior of the second vehicle. However, the second vehicle behavior estimation unit 52 may estimate various vehicle behaviors of the second vehicle other than the probability of the second vehicle changing lanes, such as deceleration, acceleration, stopping, etc.

[0099] Although various exemplary embodiments and examples are described in this disclosure, the various features, aspects, and functions described in one or more embodiments are not limited to the application of a particular embodiment, but may be applied to the embodiments alone or in various combinations. Therefore, countless variations not illustrated are contemplated within the scope of the technology disclosed in this disclosure specification. For example, this includes cases where at least one component is modified, added, or omitted, or where at least one component is extracted and combined with components of another embodiment.

[0100] 1: Vehicle behavior estimation device, 51: Information acquisition unit, 52: Second vehicle behavior estimation unit, 53: Reliability calculation unit, 54: Vehicle behavior determination unit, 55: Vehicle control unit

Claims

1. A vehicle behavior estimation device comprising: an information acquisition unit that acquires information about the road along which a first vehicle is traveling, and information about surrounding vehicles present around the first vehicle; a second vehicle behavior estimation unit that sets a second vehicle as a subject of vehicle behavior estimation from the surrounding vehicles, and estimates the vehicle behavior of the second vehicle based on at least the information about the second vehicle and the road information; and a reliability calculation unit that calculates an estimation reliability of the vehicle behavior of the second vehicle according to a degree of difficulty in estimating the vehicle behavior of the second vehicle, based on at least one of the information about the first vehicle, the information about the second vehicle, and the road information.

2. A vehicle behavior estimation device as described in claim 1, wherein the second vehicle behavior estimation unit calculates information regarding the lateral movement of the second vehicle based on information about the second vehicle and the road information, and estimates the probability of a lane change due to the lateral movement of the second vehicle as the vehicle behavior of the second vehicle based on the information regarding the lateral movement.

3. The vehicle behavior estimation device of claim 1, wherein the second vehicle behavior estimation unit calculates information regarding the lateral movement of the second vehicle based on information about the second vehicle and the road information, estimates the probability of a lane change due to the lateral movement of the second vehicle based on the information regarding the lateral movement, estimates the probability of a lane change due to an inducing factor that induces a lane change of the second vehicle due to a relationship between the second vehicle and the surrounding environment of the second vehicle, and calculates the probability of a lane change of the second vehicle as the vehicle behavior of the second vehicle by taking a weighted average of the probability of a lane change due to the lateral movement and the probability of a lane change due to the inducing factor.

4. A vehicle behavior estimation device as described in claim 2 or 3, wherein the second vehicle behavior estimation unit calculates, as information regarding the lateral movement of the second vehicle, at least one piece of information including the lateral position, speed, acceleration, amount of change in position over time, amount of change in speed over time, and amount of change in acceleration over time of the second vehicle in relation to the lane in which the second vehicle is traveling, and based on the calculated information regarding the lateral movement, the greater the tendency of the second vehicle to move laterally relative to the lane in which the second vehicle is traveling, the higher the estimated probability of a lane change due to the lateral movement.

5. A vehicle behavior estimation device as described in any one of claims 2 to 4, wherein the second vehicle behavior estimation unit determines the probability of a lane change due to lateral movement of the second vehicle based on information regarding the lateral movement of the second vehicle using a learning model trained to input information regarding the lateral movement of the second vehicle and output the probability of the lane change due to the lateral movement.

6. A vehicle behavior estimation device as described in any one of claims 1 to 5, wherein the reliability calculation unit calculates the estimation reliability based on at least one of the following factors contributing to the estimation difficulty: the distance between the first vehicle and the second vehicle, the road curvature related to the detection of the second vehicle, the road gradient related to the detection of the second vehicle, and the degree of overlap between the second vehicle and other vehicles related to the detection of the second vehicle.

7. A vehicle behavior estimation device as described in any one of claims 1 to 6, wherein the reliability calculation unit calculates the estimation reliability based on at least the detection state of a sensor detecting the second vehicle as a factor contributing to the estimation difficulty, and the detection state of the sensor includes one or more of reliability information of the detection result output from each sensor detecting the second vehicle, the number of sensors among the multiple sensors for detecting the second vehicle whose detection results can be used, the presence or absence of outliers in the detection values ​​of the multiple sensors for detecting the second vehicle, and the presence or absence of temporal discontinuity in the detection values ​​of the multiple sensors for detecting the second vehicle.

8. A vehicle behavior estimation device as described in any one of claims 1 to 7, wherein the reliability calculation unit reduces the estimation reliability when the past estimation result of the vehicle behavior by the second vehicle behavior estimation unit is incorrect.

9. A vehicle behavior estimation device as described in any one of claims 1 to 8, further comprising a vehicle behavior judgment unit that makes a final judgment as to whether or not the second vehicle will execute the estimated vehicle behavior based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit and the estimation reliability calculated by the reliability calculation unit.

10. A vehicle behavior estimation device as described in claim 9, wherein the vehicle behavior determination unit makes a final determination as to whether or not the second vehicle will execute the estimated lane change based on the probability of the second vehicle changing lanes estimated as the vehicle behavior of the second vehicle by the second vehicle behavior estimation unit and the estimation reliability calculated by the reliability calculation unit.

11. A vehicle behavior estimation device as described in any one of claims 1 to 10, further comprising a vehicle control unit that calculates a control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit and the estimation reliability calculated by the reliability calculation unit.

12. A vehicle behavior estimation device as described in claim 11, wherein the vehicle control unit calculates the control amount of the first vehicle based on the vehicle behavior of the second vehicle estimated by the second vehicle behavior estimation unit, and changes the magnitude of the calculated control amount of the first vehicle based on the estimation reliability calculated by the reliability calculation unit.

13. A vehicle behavior estimation method comprising: an information acquisition step of acquiring information about a road along which a first vehicle is traveling, and information about surrounding vehicles present around the first vehicle; a second vehicle behavior estimation step of setting a second vehicle as a subject of vehicle behavior estimation from the surrounding vehicles, and estimating the vehicle behavior of the second vehicle based on at least the information about the second vehicle and the road information; and a reliability calculation step of calculating an estimation reliability of the vehicle behavior of the second vehicle according to a degree of difficulty in estimating the vehicle behavior of the second vehicle, based on at least one of the information about the first vehicle, the information about the second vehicle, and the road information.

Images