FOREIGN VEHICLE BEHAVIOR PREDICTION DEVICE, FOREIGN VEHICLE BEHAVIOR PREDICTION METHOD AND NON-FLYING RECORDING PORT

Abstract

A foreign vehicle behavior prediction device, based on the acquisition results of an environment situation sensor for sensing the environment situation of the host vehicle, predicts whether a driver is in a foreign vehicle located in the vicinity of the host vehicle, and, based on the prediction results of whether the driver is in the foreign vehicle and the time series acquisition results of the foreign vehicle by the environment situation sensor, predicts the behavior of the foreign vehicle.

Description

FIELD The present disclosure relates to a foreign vehicle behavior prediction device, a foreign vehicle behavior prediction method and a non-volatile recording medium. BACKGROUND PTL 1 ( JP 2023-508986 A ) describes a technology for predicting a user's intention to share a road with a vehicle. Although the prediction of the behavior of an unrelated vehicle (another vehicle) has been performed in the past, the behavior of the unrelated vehicle was predicted without considering whether a driver was present in the unrelated vehicle. It is assumed that an unrelated vehicle manually controlled by a driver frequently performs lane changes, accelerations, decelerations, or similar actions, even in situations where there are no intersections, no vehicles traveling at low speeds ahead, or similar situations. In contrast, it is assumed that an unrelated vehicle autonomously controlled without a driver rarely performs lane changes, accelerations, decelerations, or similar actions in situations where there are no intersections, no vehicles traveling at low speeds ahead, or similar situations.However, since the behavior of the foreign vehicle was predicted in the past without taking into account whether there was a driver in the foreign vehicle, the behavior of the foreign vehicle cannot be predicted with high accuracy. SUMMARY In view of the foregoing, an objective of the present disclosure is to provide a foreign vehicle behavior prediction device, a foreign vehicle behavior prediction method and a non-volatile recording medium with which the behavior of the foreign vehicle (the other vehicle) can be predicted with high accuracy.(1) One embodiment of the present disclosure relates to a third-party vehicle behavior prediction device comprising a processor configured to: predict, based on the detection results of an environment situation sensor detecting the environment situation of the host vehicle, whether a driver is in a third-party vehicle located in the vicinity of the host vehicle; and, based on the prediction results of whether the driver is in the third-party vehicle and the time-series detection results of the third-party vehicle by the environment situation sensor, predict the behavior of the third-party vehicle.(2) In the foreign vehicle behavior prediction device according to embodiment (1), the behavior of the foreign vehicle that is predicted when it is predicted that the driver is in the foreign vehicle and the behavior of the foreign vehicle that is predicted when it is predicted that the driver is not in the foreign vehicle may be different.(3) In the foreign vehicle behavior prediction device according to embodiment (1) or (2), the processor can be configured to predict the possibility that the foreign vehicle will perform a lane change, based on the prediction results regarding whether the driver is present in the foreign vehicle and the time series recording results of the foreign vehicle by the environmental situation sensor, by using a prediction model obtained by learning using teacher data, which is a dataset of time series recording results of a learning foreign vehicle from a first time point to a second time point by a learning environmental situation sensor attached to a learning host vehicle, and labels indicating information about whether a driver is present in the learning foreign vehicle.Time series recording results of a learning foreign vehicle from a first time point to a second time point by a learning environment situation sensor attached to a learning host vehicle, and labels indicating whether a driver is present in the learning foreign vehicle, and information indicating whether the learning foreign vehicle changed lanes at a third time point, which is later than the second time point, were obtained.(4) In the foreign vehicle behavior prediction device according to one of embodiments (1) to (3), the processor can be configured to predict, based on the behavior of the foreign vehicle detected by the environment situation sensor, whether a driver is present in the foreign vehicle.(5) In the foreign vehicle behavior prediction device according to any embodiment (1) to (4), the processor may be configured to predict, on the basis of an image of the foreign vehicle taken by a camera serving as an environmental situation sensor, whether the driver is present in the foreign vehicle. (6) In the foreign vehicle behavior prediction device according to any embodiment (1) to (5), the processor may be configured to predict, on the basis of information indicating whether the driver is present in the foreign vehicle, which is acquired via wireless communication with a location outside the host vehicle by means of a wireless communication device serving as an environmental situation sensor, whether the driver is present in the foreign vehicle.(7) One embodiment of the present disclosure relates to a foreign vehicle behavior prediction method comprising: predictions as to whether a driver is in a foreign vehicle located in the vicinity of a host vehicle, based on the detection results of an environmental situation sensor for detecting the environmental situation of the host vehicle; and predictions of the behavior of the foreign vehicle based on the environmental situation prediction results of whether the driver is in the foreign vehicle and the time series detection results of the foreign vehicle by the environmental situation sensor.(8) One embodiment of the present disclosure relates to a non-volatile recording medium on which a computer program is recorded that causes a processor to execute a process comprising: predicting whether a driver is in a foreign vehicle located in the vicinity of a host vehicle, based on the detection results of an environmental situation sensor for detecting the environmental situation of the host vehicle; and predicting the behavior of the foreign vehicle based on prediction results as to whether the driver is present in the foreign vehicle and time-series detection results of the foreign vehicle by the environmental situation sensor. According to the present disclosure, the behavior of the foreign vehicle can be predicted with high accuracy. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing an example of a host vehicle 1 in which a foreign vehicle behavior prediction device 15 is applied according to a first embodiment. Fig. 2A is a view showing an example of prediction results of the behavior of a foreign vehicle OV by a foreign vehicle behavior prediction unit 3D when a foreign vehicle driver prediction unit 3C predicts that there is a driver in the foreign vehicle OV, and so on. Fig. 2B is a view showing an example of the prediction results of the behavior of the foreign vehicle OV by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that there is no driver in the foreign vehicle OV, and so on.Figure 3 is a flowchart illustrating an example of a process performed by a processor 153 of the foreign vehicle behavior prediction device 15 of the first embodiment. Figure 4A is a view showing an example of the foreign vehicle behavior prediction results by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that a driver is in the foreign vehicle OV, and so on. Figure 4B is a view showing an example of the foreign vehicle behavior prediction results by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that there is no driver in the foreign vehicle OV, and so on. DESCRIPTION OF EXAMPLES OF EXECUTION The embodiments of the device for predicting the behavior of the foreign vehicle, the method for predicting the behavior of the foreign vehicle and the non-volatile recording medium of the present disclosure are described below with reference to the drawings. <Erstes Ausführungsbeispiel> Fig. 1 is a view showing an example of a host vehicle 1 in which a foreign vehicle behavior prediction device 15 of a first embodiment is applied. In the example shown in Fig. 1, the host vehicle 1 comprises an environment situation sensor 11, a vehicle condition sensor 12, a human-machine interface (HMI) 13, a vehicle control device 14, a steering actuator 14A, a brake actuator 14B, a drive actuator 14C and the foreign vehicle behavior prediction device 15. The environmental situation sensor 11 detects the environmental situation of the host vehicle 1 (e.g., other vehicles OV (see Fig. 2A and Fig. 2B) located in the vicinity of the host vehicle 1, obstacles located in the vicinity of the host vehicle 1, etc.) and transmits the detection results of the environmental situation of the host vehicle 1 to the vehicle control unit 14 and the other vehicle behavior prediction device 15. The environmental situation sensor 11 includes, for example, a camera, a LiDAR (Light Detection and Ranging) system, a radar system, a wireless communication device for acquiring information representing a situation outside the host vehicle 1 from outside the host vehicle 1 via wireless communication, etc. The vehicle condition sensor 12 performs a state detection of the host vehicle 1, a measurement of the host vehicle 1's position, and the like, and transmits the state detection results, the position measurement results, and the like to the vehicle control device 14 and the other vehicle behavior prediction device 15. The vehicle condition sensor 12 includes, for example, a vehicle speed sensor, an accelerometer, a yaw rate sensor, a gyroscope, a GPS receiver (Global Positioning System), etc. The HMI 13 has the function of accepting various operating operations by the driver of the host vehicle 1 and the like, and transmits signals representing the operating operations by the driver of the host vehicle 1 to the vehicle control unit 14. The vehicle control unit 14 controls the steering actuator 14A, the brake actuator 14B, and the drive actuator 14C, for example, based on information (data, signals), etc., transmitted by the environmental situation sensor 11, the vehicle status sensor 12, and the HMI 13. Specifically, the vehicle control unit 14 has an autonomous driving function to control the steering actuator 14A, the brake actuator 14B, and the drive actuator 14C so that the host vehicle 1 drives autonomously without requiring operation by the driver of the host vehicle 1. In particular, the vehicle control unit 14 generates a route plan for the host vehicle 1 to reach its destination, based on, for example, map information, position information of the host vehicle 1, information representing the destination of the host vehicle 1, etc.Furthermore, the vehicle control device 14 instructs the host vehicle 1 to drive autonomously according to the schedule. Specifically, the vehicle control device 14 instructs the host vehicle 1 to drive autonomously while revising the schedule to avoid collisions between the host vehicle 1 and the foreign vehicle OV (see Fig. 2A and Fig. 2B), etc., based on the detection results of the environmental situation sensor 11, the predictive results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction device 15, which will be described later, etc. The foreign vehicle behavior prediction device 15 consists of a microcomputer with a communication interface (I / F) 151, a memory 152 and a processor 153. The communication interface 151 comprises an interface circuit for connecting the foreign vehicle behavior prediction device 15 with the environmental situation sensor 11, the vehicle status sensor 12, the HMI 13, and the vehicle control device 14. The memory 152 stores a program used in a process executed by the processor 153, as well as various data. The processor 153 has a function as a procurement unit 3A, a function as an object lane detection unit 3B, a function as a foreign vehicle driver prediction unit 3C, and a function as a foreign vehicle behavior prediction unit 3D. Procurement unit 3A records the results of the environmental conditions of host vehicle 1 and the measurement results of the position of host vehicle 1. The object lane detection unit 3B performs the detection of an object such as the foreign vehicle OV (see Fig. 2A and Fig. 2B) or the like, which is in the vicinity of the host vehicle 1, and the detection of lanes that are in the vicinity of the host vehicle 1, based on the detection results of the environmental situation of the host vehicle 1 acquired by the procurement unit 3A. Based on the acquisition results (sensor data from the environmental situation sensor 11) of the environmental situation of the host vehicle 1 acquired by the procurement unit 3A, the foreign vehicle driver prediction unit 3C predicts whether there is a driver in the foreign vehicle OV that is in the vicinity of the host vehicle 1. In the example shown in Fig. 1, the foreign vehicle driver prediction unit 3C uses the behavior of the foreign vehicle OV (time-series sensor data from the environmental situation sensor 11), which is detected, for example, by the camera or a similar device serving as the environmental situation sensor 11, to predict whether the driver is in the foreign vehicle OV. Specifically, if the environmental situation sensor 11 detects human-specific driving behavior of the foreign vehicle OV, such as sudden acceleration, sudden braking, frequent lane changes, etc., the foreign vehicle driver prediction unit 3C predicts that the driver is present in the foreign vehicle OV. In another example, the foreign vehicle driver prediction unit 3C predicts whether the driver is in the foreign vehicle OV (see Fig. 2A and Fig. 2B) based on an image of the foreign vehicle OV captured by the camera acting as an environmental situation sensor 11. For example, if the driver of the foreign vehicle OV is included in the image containing a rearview mirror or a side mirror of the foreign vehicle OV captured by the camera acting as an environmental situation sensor 11 (especially if the driver of the foreign vehicle OV is reflected in the rearview or side mirror of the foreign vehicle OV), the foreign vehicle driver prediction unit 3C predicts that the driver is present in the foreign vehicle OV. In another example, the foreign vehicle driver prediction unit 3C predicts whether the driver is present in the foreign vehicle OV based on information indicating the driver's presence in the foreign vehicle OV, which is acquired from outside the host vehicle 1 by a wireless communication device acting as an environmental situation sensor 11. Specifically, the wireless communication device acting as an environmental situation sensor 11 acquires the information indicating the driver's presence in the foreign vehicle OV by, for example, performing V2I (vehicle-to-roadside infrastructure) or V2V (vehicle-to-vehicle) communication, and the foreign vehicle driver prediction unit 3C uses this information to predict whether the driver is present in the foreign vehicle OV. In the example shown in Fig. 1, the foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the prediction results of the foreign vehicle driver prediction unit 3C, whether the driver is present in the foreign vehicle OV, and the time series recording results DR (see Fig. 2A and Fig. 2B) of the foreign vehicle OV by the environmental situation sensor 11. Figures 2A and 2B are views showing examples of the time-series acquisition results DR of the foreign vehicle OV by the environmental situation sensor 11 and the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D. Specifically, Figure 2A shows an example of the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that the driver is present in the foreign vehicle OV, and Figure 2B shows an example of the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that the driver is not present in the foreign vehicle OV. In the example shown in Fig. 2A, the host vehicle 1 travels in lane L1 and the other vehicle OV travels in lane L2. Specifically, the other vehicle OV passes a position P2 of lane L2. The other vehicle behavior prediction device 15 predicts the behavior of the other vehicle OV after the time shown in Fig. 2A so that the host vehicle 1 can travel safely without a collision or similar occurrence between the host vehicle 1 and the other vehicle OV. In particular, in the example shown in Fig. 2A, the foreign vehicle driver prediction unit 3C predicts, based on the detection results of the environment situation sensor 11, that the driver is in the foreign vehicle OV (in particular, there is the possibility that the foreign vehicle OV is being driven manually). The foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the prediction results that the driver is in the foreign vehicle OV, which were determined by the foreign vehicle driver prediction unit 3C and the time series acquisition results DR of the foreign vehicle OV. (More precisely, the positional trajectory of the foreign vehicle OV from the time the foreign vehicle OV passes a position P1 until the time the foreign vehicle OV passes a position P2) by the environmental situation sensor 11. Specifically, the foreign vehicle behavior prediction unit 3D predicts that the probability of the foreign vehicle OV changing from lane L2 to lane L3 is 30%, that the probability of the foreign vehicle OV continuing on lane L2 without changing lanes is 60%, and that the probability of the foreign vehicle OV changing from lane L2 to lane L1 is 10%. In the example shown in Fig. 2B, the host vehicle 1 is traveling in lane L1 and the other vehicle OV is traveling in lane L2. More precisely, the other vehicle OV passes position P2 of lane L2. The other vehicle behavior prediction device 15 predicts the behavior of the other vehicle OV after the time shown in Fig. 2B so that the host vehicle 1 can travel safely without a collision or similar incident occurring between the host vehicle 1 and the other vehicle OV. In particular, in the example shown in Fig. 2B, the foreign vehicle driver prediction unit 3C predicts, based on the detection results of the environment situation sensor 11, that the driver is not present in the foreign vehicle OV (i.e., the foreign vehicle OV is driving autonomously). The foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the prediction results that the driver is not present in the foreign vehicle OV, which are derived from the foreign vehicle driver prediction unit 3C and the time series acquisition results DR of the foreign vehicle OV. (More precisely, the positional trajectory of the foreign vehicle OV from the time the foreign vehicle OV passes position P1 until the time the foreign vehicle OV passes position P2) by the environmental situation sensor 11. Specifically, the foreign vehicle behavior prediction unit 3D predicts that the probability of foreign vehicle OV changing from lane L2 to lane L3 is 10%, that the probability of foreign vehicle OV continuing on lane L2 without changing lanes is 80%, and that the probability of foreign vehicle OV changing from lane L2 to lane L1 is 10%. In the examples shown in Fig. 2A and Fig. 2B, the behavior (the probability that the foreign vehicle OV changes from lane L2 to lane L3 is 30%, the probability that the foreign vehicle OV continues in lane L2 without changing lanes is 60%, and the probability that the foreign vehicle OV changes from lane L2 to lane L1 is 10%) of the foreign vehicle OV, as predicted by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that the driver is in the foreign vehicle OV, and the behavior (the probability that the foreign vehicle OV changes from lane L2 to lane L3 is 10%, the probability that the foreign vehicle OV continues in lane L2 without changing lanes is 80%, and the probability that the foreign vehicle OV changes from lane L2 to lane L1 is 10%) is 10%. changesThe percentage of the foreign vehicle OV predicted by the foreign vehicle behavior prediction unit 3D (10%) when the foreign vehicle driver prediction unit 3C predicts that the driver is not present in the foreign vehicle OV) differs. In the examples shown in Fig. 2A and Fig. 2B, the position path of the foreign vehicle OV from the time at which the foreign vehicle OV passes position P1 until the time at which the foreign vehicle OV passes position P2 is used as a time series recording result DR of the foreign vehicle OV by the environmental situation sensor 11; in another example, the recording results of the orientation of the foreign vehicle OV from the time at which the foreign vehicle OV passes position P1 until the time at which the foreign vehicle OV passes position P2 can be used as time series recording results DR of the foreign vehicle OV by the environmental situation sensor 11. In another example, the time series recording results DR of the foreign vehicle OV by the environmental situation sensor 11, the recording results such as speed, acceleration, braking behavior, etc. of the foreign vehicle OV from the time at which the foreign vehicle OV passes position P1, until the time at which the foreign vehicle OV passes position P2, can be used. In the example shown in Fig. 1, the stranger vehicle behavior prediction unit 3D predicts the probability that the stranger vehicle OV will perform a lane change, based on the prediction results of whether the driver is present in the stranger vehicle OV by the stranger vehicle driver prediction unit 3C and the time series acquisition results of the stranger vehicle OV by the environmental situation sensor 11, using a prediction model that is learned by training using teacher data, which is a dataset of time series acquisition results of a learning stranger vehicle (not shown) from a first time point to a second time point by a learning environmental situation sensor (not shown) attached to a learning host vehicle (not shown), and labels that display information about it.whether time series recording results of a learning foreign vehicle (not shown) from a first time point to a second time point by a learning environment situation sensor (not shown) attached to a learning host vehicle (not shown), and labels indicating whether the driver is present in the learning foreign vehicle, and information on whether the learning foreign vehicle changed lanes at a third time point, which is later than the second time point. In another example, the foreign vehicle behavior prediction unit 3D can predict the probability that the foreign vehicle OV will perform the lane change, based on the prediction results of whether the driver is present in the foreign vehicle OV, by the foreign vehicle driver prediction unit 3C and the time series recording results of the foreign vehicle OV by the environmental situation sensor 11, using a prediction model obtained by a different method than the example shown in Fig. 1. Fig. 3 is a flowchart to illustrate an example of the process carried out by the processor 153 of the foreign vehicle behavior prediction device 15 of the first embodiment. In the example shown in Fig. 3, the procurement unit 3A records the recording results of the environmental situation of the host vehicle 1 and the measurement results of the position of the host vehicle 1 in step S10. In step S11, the object lane detection unit 3B performs the detection of the object, such as the foreign vehicle OV, which is in the vicinity of the host vehicle 1, and the detection of the lanes which are in the vicinity of the host vehicle 1, based on the detection results of the environmental situation of the host vehicle 1 acquired in step S10. In step S12, the foreign vehicle driver prediction unit 3C, based on the environmental situation data collected in step S10 for host vehicle 1, predicts whether a driver is in the foreign vehicle OV located in the vicinity of host vehicle 1. If YES, the process proceeds to step S13, and if NO, the process proceeds to step S14. In step S13, the foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the time series acquisition results DR of the foreign vehicle OV by the environmental situation sensor 11 and the driving behavior specific to humans (the driver of the foreign vehicle OV) (irregular driving behavior based on emotions, attention, experience, etc. of the human). In step S14, the foreign vehicle behavior prediction unit 3D predicts the behavior (regular and safe driving behavior according to a programmed algorithm) of the foreign vehicle OV based on the time series recording results DR of the foreign vehicle OV by the environmental situation sensor 11 and the properties (algorithm) of the AI ​​(artificial intelligence) applied to the foreign vehicle OV (autonomous vehicle). As described above, in the foreign vehicle behavior prediction device 15 of the first embodiment, in contrast to the prior art, where the behavior of the foreign vehicle OV is predicted under the assumption that the foreign vehicle OV is manually controlled by the driver of the foreign vehicle OV, the behavior of the autonomously controlled vehicle (other vehicle OV), for which the prediction based on human reactions and driving habits does not apply, is predicted with high accuracy in the case of the foreign vehicle OV being an autonomously controlled vehicle. This allows the safety and reliability of the autonomous driving of the host vehicle 1 to be improved. <Zweites Ausführungsbeispiel> The host vehicle 1, to which the foreign vehicle behavior prediction device 15 of a second embodiment is applied, is configured in the same way as the host vehicle 1 to which the foreign vehicle behavior prediction device 15 of the first embodiment is applied, except for the points described below. Figures 4A and 4B are views showing examples of the time series acquisition results DR of the foreign vehicle OV by the environmental situation sensor 11 and the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D of the foreign vehicle behavior prediction device 15 of the second embodiment. Specifically, Figure 4A shows an example of the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that the driver is present in the foreign vehicle OV, and Figure 4B shows an example of the prediction results of the foreign vehicle OV's behavior by the foreign vehicle behavior prediction unit 3D when the foreign vehicle driver prediction unit 3C predicts that there is no driver in the foreign vehicle OV. In the example shown in Fig. 4A, the host vehicle 1 is traveling in lane L2, and the foreign vehicle OV is traveling in front of the host vehicle 1. More precisely, the foreign vehicle OV passes position P2 of lane L2. The foreign vehicle behavior prediction device 15 predicts the behavior of the foreign vehicle OV after the time shown in Fig. 4A so that the host vehicle 1 can travel safely without a collision or similar incident occurring between the host vehicle 1 and the foreign vehicle OV. In particular, in the example shown in Fig. 4A, the foreign vehicle driver prediction unit 3C predicts, based on the detection results of the environment situation sensor 11, that a driver is in the foreign vehicle OV (i.e., that there is a possibility that the foreign vehicle OV is being driven manually). The foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the prediction results that the driver is in the foreign vehicle OV, provided by the foreign vehicle driver prediction unit 3C, and the time series acquisition results DR of the foreign vehicle OV (more precisely, the speed, acceleration, braking behavior, etc.) from the time the foreign vehicle OV passes position P1 until the time the foreign vehicle OV passes position P2, provided by the environmental situation sensor 11. Specifically, the foreign vehicle behavior prediction unit 3D predicts that the probability that the foreign vehicle OV will accelerate is 20%, that the probability that the foreign vehicle OV will continue at a constant speed without accelerating or decelerating is 60%, and that the probability that the foreign vehicle OV will decelerate is 20%. In the example shown in Fig. 4B, the host vehicle 1 is traveling in lane L2, and the foreign vehicle OV is traveling in front of the host vehicle 1. More precisely, the foreign vehicle OV passes position P2 of lane L2. The foreign vehicle behavior prediction device 15 predicts the behavior of the foreign vehicle OV after the time shown in Fig. 4B so that the host vehicle 1 can travel safely without a collision or similar incident occurring between the host vehicle 1 and the foreign vehicle OV. In particular, in the example shown in Fig. 4B, the foreign vehicle driver prediction unit 3C predicts, based on the detection results of the environment situation sensor 11, that there is no driver in the foreign vehicle OV (i.e., the foreign vehicle OV is driving autonomously). The foreign vehicle behavior prediction unit 3D predicts the behavior of the foreign vehicle OV based on the prediction results that there is no driver in the foreign vehicle OV, provided by the foreign vehicle driver prediction unit 3C, and the time series acquisition results DR of the foreign vehicle OV (more precisely, the speed, acceleration, braking behavior, etc.) from the time the foreign vehicle OV passes position P1 until the time the foreign vehicle OV passes position P2, provided by the environmental situation sensor 11. Specifically, the foreign vehicle behavior prediction unit 3D predicts that the probability that the foreign vehicle OV will accelerate is 10%, that the probability that the foreign vehicle OV will continue at a constant speed without accelerating or decelerating is 80%, and that the probability that the foreign vehicle OV will decelerate is 10%. In an example for the host vehicle 1, to which the foreign vehicle behavior prediction device 15 of the second embodiment is applied, the foreign vehicle behavior prediction unit 3D predicts the probability that the foreign vehicle OV will accelerate or decelerate, based on the prediction results of the foreign vehicle driver prediction unit 3C as to whether the driver is in the foreign vehicle OV, and the time series acquisition results of the detection of the foreign vehicle OV by the environment situation sensor 11 using a prediction model obtained by learning using teacher data, which is a dataset of time series acquisition results of the detection of a learning foreign vehicle (not shown) from a first time point to a second time point by a learning environment situation sensor (not shown) attached to a learning host vehicle (not shown).and labels containing information on whether the driver is present in the learner vehicle, and information on whether the learner vehicle accelerated or decelerated at a third time later than the second time. In another example, the foreign vehicle behavior prediction unit 3D can predict the possibility that the foreign vehicle (other vehicle) OV will accelerate or decelerate, based on the prediction results of whether the driver is present in the foreign vehicle OV by the foreign vehicle driver prediction unit 3C and the time series acquisition results of the foreign vehicle OV by the environment situation sensor 11, using a prediction model obtained by a different method than the example described above. <Drittes Ausführungsbeispiel> The host vehicle 1, to which the foreign vehicle behavior prediction device 15 of a third embodiment is applied, is configured in the same way as the host vehicle 1, to which the foreign vehicle behavior prediction device 15 of the first or second embodiment is applied, except for the points described below. As described above, the vehicle control device 14 in the host vehicle 1, to which the other vehicle behavior prediction device 15 of the first embodiment is attached, has an automatic driving function for controlling the steering actuator 14A, the brake actuator 14B, and the drive actuator 14C to cause the host vehicle 1 to drive autonomously without requiring operation by the driver of the host vehicle 1. In particular, the vehicle control device 14 generates the route plan for the host vehicle 1 to reach its destination, based on, for example, map information, the position information of the host vehicle 1, information representing the destination of the host vehicle 1, etc. Furthermore, the vehicle control device 14 causes the host vehicle 1 to drive autonomously according to the route plan.Specifically, the vehicle control unit 14 causes the host vehicle 1 to drive autonomously while it revises the route plan to avoid collisions between the host vehicle 1 and the foreign vehicle OV (see Fig. 2A and Fig. 2B), etc., based on the detection results of the environmental situation sensor 11, the prediction results of the behavior of the foreign vehicle OV by the foreign vehicle behavior prediction device 15, etc. In contrast, the vehicle control unit 14 in the host vehicle 1, to which the other vehicle behavior prediction device 15 of the third embodiment is applied, has a driver assistance function. In particular, if, based on the detection results of the environmental situation sensor 11 and the prediction results of the other vehicle OV's behavior by the other vehicle behavior prediction device 15, etc., it is predicted that a collision between the host vehicle 1 and the other vehicle OV (see Fig. 2A and Fig. 2B) must be avoided, the vehicle control unit 14 causes the HMI 13 to issue a corresponding warning message. Although the embodiments of the foreign vehicle behavior prediction device, the foreign vehicle behavior prediction method, and the non-volatile recording medium of this disclosure have been explained with reference to the drawings described above, the foreign vehicle behavior prediction device, the foreign vehicle behavior prediction method, and the non-volatile recording medium of this disclosure are not limited to the embodiments described above, and suitable modifications can be made without departing from the basic concept of this disclosure. The configurations of the examples of the embodiments described above can be combined appropriately.Although the process performed by the foreign vehicle behavior prediction device 15 in each of the embodiments described above has been described as a software process carried out by executing the program, the process performed by the foreign vehicle behavior prediction device 15 can also be a hardware-based process. Alternatively, the process performed by the foreign vehicle behavior prediction device 15 can also be a process that combines both software and hardware. Furthermore, the program stored in the memory 152 of the foreign vehicle behavior prediction device 15 (the program for implementing the functions of the processor 153 of the foreign vehicle behavior prediction device 15) can be stored on a computer-readable storage medium (non-volatile recording medium) such as a semiconductor memory, a magnetic recording medium, an optical recording medium, etc.recorded, provided, distributed, etc. QUOTES INCLUDED IN THE DESCRIPTION This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature JP 2023-508986 A

[0002]

Claims

Foreign vehicle behavior prediction device with a processor configured to predict whether a driver is present in a foreign vehicle positioned in the vicinity of a host vehicle, based on acquisition results from an environmental situation sensor for acquiring an environmental situation of the host vehicle, and based on prediction results regarding whether the driver is present in the foreign vehicle and time series acquisition results of the foreign vehicle by the environmental situation sensor, to predict the behavior of the foreign vehicle. Foreign vehicle behavior prediction device according to claim 1, wherein the behavior of the foreign vehicle that is predicted when it is predicted that the driver is present in the foreign vehicle and the behavior of the foreign vehicle that is predicted when it is predicted that the driver is not present in the foreign vehicle are different. Foreign vehicle behavior prediction device according to claim 1, wherein the processor is configured, based on the prediction results regarding whether the driver is present in the foreign vehicle and the time series acquisition results of the foreign vehicle by the environmental situation sensor, by using a prediction model that is obtained by performing a learning process using teacher data, which is a dataset of time series acquisition values ​​of a learning foreign vehicle from a first time point to a second time point, by a learning environmental situation sensor mounted on a learning host vehicle, and labels indicating whether a driver is present in the learning foreign vehicle and information regarding whether the learning foreign vehicle changed lanes at a third time point, which is later than the second time point, to predict a possibility ofthat the other vehicle is changing lanes. Foreign vehicle behavior prediction device according to claim 1, wherein the processor is configured to predict, based on the behavior of the foreign vehicle detected by the environmental situation sensor, whether the driver is present in the foreign vehicle. Foreign vehicle behavior prediction device according to claim 1, wherein the processor is configured to predict, on the basis of an image of the foreign vehicle taken by a camera serving as the environment situation sensor, whether the driver is present in the foreign vehicle. Foreign vehicle behavior prediction device according to claim 1, wherein the processor is configured to predict, based on the behavior of the foreign vehicle detected by the environmental situation sensor, whether the driver is present in the foreign vehicle. Foreign vehicle behavior prediction method with: predictions of whether a driver is present in a foreign vehicle positioned in the vicinity of a host vehicle, based on acquisition results from an environmental situation sensor for acquiring an environmental situation of the host vehicle, and predictions of foreign vehicle behavior based on prediction results regarding whether the driver is present in the foreign vehicle and time series acquisition results of the foreign vehicle by the environmental situation sensor. Non-volatile recording medium containing a computer program recorded on it to cause a processor to execute a process comprising: predictions as to whether a driver is present in a foreign vehicle positioned in the vicinity of a host vehicle, based on detection results from an environmental situation sensor for detecting an environmental situation of the host vehicle; and predictions of the foreign vehicle's behavior based on prediction results regarding whether the driver is present in the foreign vehicle and time-series detection results of the foreign vehicle by the environmental situation sensor.

Images