JP7277202B2 - Vehicle behavior prediction method, vehicle behavior prediction device, and vehicle control device - Google Patents

Description

The present invention relates to a vehicle behavior prediction method, a vehicle behavior prediction device, and a vehicle control device.

Conventionally, there has been known a technique for determining a lane change of a vehicle (for example, Patent Document 1). The technology described in Patent Document 1 detects another vehicle traveling in an adjacent lane, obtains the angle of the movement vector of the other vehicle, and if the angle of the movement vector satisfies a predetermined condition, detects the vehicle from the adjacent lane. It is determined that other vehicles will change lanes. In addition, when it is determined that another vehicle will cut in from the adjacent lane, this technology predicts the distance between the other vehicle and the own vehicle when the other vehicle changes lanes, and predicts the distance between the vehicles and the target. Vehicle speed control is performed so that the inter-vehicle distance matches.

Japanese Patent Application Laid-Open No. 2001-199260

However, with the technique described in Patent Literature 1, the lane change cannot be determined unless the other vehicle actually starts changing lanes and the movement vector does not meet a certain condition. Therefore, there is a problem that it is not possible to predict lane changes of other vehicles at an early stage.

SUMMARY OF THE INVENTION The present invention has been made in view of such problems, and an object of the present invention is to provide a vehicle behavior prediction method, a vehicle behavior prediction device, and a vehicle control device that can predict lane changes of other vehicles at an early stage. be.

A vehicle behavior prediction method according to an aspect of the present invention specifies a first other vehicle for which the possibility of changing lanes is predicted, and acquires traffic conditions ahead of a second other vehicle that stops ahead of the first other vehicle. Then , based on the map information and the traffic condition, a second stop factor, which is the cause of the stop of the other vehicle, is obtained. The vehicle behavior prediction method predicts the timing at which the stop factor is resolved, and predicts the predicted start time, which is the time until the second other vehicle starts, based on the timing at which the stop factor is resolved . Then, the vehicle behavior prediction method predicts that the longer the predicted start time, the higher the possibility that the first other vehicle will change lanes.

According to the present invention, lane changes of other vehicles can be predicted early.

FIG. 1 is a block diagram showing the configuration of a vehicle behavior prediction device according to this embodiment. FIG. 2 is a flowchart showing the vehicle behavior prediction processing procedure according to the present embodiment. FIG. 3A is an explanatory diagram schematically showing an intersection, which is an example of an intersection position. FIG. 3B is an explanatory diagram showing a position where side roads intersect, which is an example of the intersection position. FIG. 3C is an explanatory diagram showing a position where private land is adjacent, which is an example of the intersection position. FIG. 4 is a flow chart showing a processing procedure for predicting the predicted start time. FIG. 5 is a diagram schematically showing traffic conditions at an intersection. FIG. 6 is a flow chart showing a processing procedure for predicting the possibility of lane change. FIG. 7A is an explanatory diagram showing the relationship between the predicted start time and the possibility of changing lanes. FIG. 7B is an explanatory diagram showing the relationship between stop time and lane change possibility. FIG. 7C is an explanatory diagram showing the relationship between the sum of the predicted start time and stop time and the possibility of lane change. FIG. 8 is a flow chart showing the procedure for calculating the effect of lane changes. FIG. 9A is an explanatory diagram showing reference positions set with respect to intersections. FIG. 9B is an explanatory diagram showing the reference position set for the side road. FIG. 10A is a diagram schematically showing an example of a situation in which lane change is highly difficult. FIG. 10B is a diagram schematically showing an example of a situation in which the difficulty of changing lanes is high. FIG. 10C is a diagram schematically showing an example of a situation in which lane change is difficult. FIG. 11A is a diagram schematically showing a vehicle entering a side road. FIG. 11B is a diagram schematically showing a state of a vehicle entering private property on the side of a road. FIG. 11C is a diagram schematically showing a traffic jam.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the description of the drawings, the same parts are denoted by the same reference numerals, and the description thereof is omitted.

The configuration of the vehicle behavior prediction device according to the present embodiment will be described with reference to FIG. The vehicle behavior prediction device has an object detection device 1 , a vehicle position estimation device 3 , a map acquisition device 4 and a microcomputer 50 .

The vehicle behavior prediction device may be applied to a vehicle that has an automatic driving function, or may be applied to a vehicle that does not have an automatic driving function. Also, the vehicle behavior prediction device may be applied to a vehicle capable of switching between automatic driving and manual driving. Hereinafter, the vehicle to which the vehicle behavior prediction device is applied will be referred to as own vehicle.

Automatic driving refers to, for example, a state in which at least one actuator among brake, accelerator, and steering actuators is controlled without the driver's operation. Therefore, other actuators may be operated by the passenger's operation. Further, automatic operation may be a state in which any control such as acceleration/deceleration control or lateral position control is being executed. Further, manual driving in this embodiment refers to a state in which the driver is operating the brake, accelerator, and steering, for example.

The object detection device 1 includes a plurality of object detection sensors such as a laser radar, a millimeter wave radar, and a camera mounted on the own vehicle. The object detection device 1 detects objects around the own vehicle using a plurality of object detection sensors. The object detection device 1 detects moving objects including other vehicles, motorcycles, bicycles and pedestrians, and stationary objects including parked vehicles and buildings. For example, the object detection device 1 detects the position, attitude (yaw angle), size, speed, acceleration, deceleration, and yaw rate of a moving object and a stationary object with respect to the host vehicle.

The own vehicle position estimation device 3 measures the absolute position of the own vehicle using position estimation techniques such as GPS (Global Positioning System) and odometry. The own vehicle position estimation device 3 uses a position detection sensor to measure the absolute position of the own vehicle, that is, the position, vehicle speed, acceleration, steering angle, and attitude of the own vehicle with respect to a predetermined reference point. The vehicle position estimation device 3 includes a GPS receiver, an inertial navigation system, sensors provided on the brake pedal and accelerator pedal, sensors for acquiring vehicle behavior such as a wheel speed sensor and a yaw rate sensor, a laser radar, a camera, and the like. is

The map acquisition device 4 acquires map information indicating the structure of the road on which the vehicle travels. The map information acquired by the map acquisition device 4 includes road structure information such as absolute position of lanes, connection relationship between lanes, relative positional relationship, traffic rules, road signs, and the like. The map information acquired by the map acquisition device 4 also includes information on facilities such as parking lots, gas stations, etc., which are private land on the side of the road. In addition, the map information includes traffic light information such as the position of the traffic light and the type of the traffic light. The map acquisition device 4 may own a map database storing map information, or may acquire map information from an external map data server by cloud computing. Further, the map acquisition device 4 may acquire map information using vehicle-to-vehicle communication or road-to-vehicle communication.

The microcomputer 50 predicts the behavior of other vehicles around the host vehicle based on the detection result of the object detection device 1, the estimation result of the host vehicle position estimation device 3, and the acquisition result of the map acquisition device 4. The microcomputer 50 also controls the running state of the own vehicle based on the predicted behavior of the other vehicle.

Behavior prediction of other vehicles according to the present embodiment is performed by predicting the behavior of another vehicle traveling in a lane (adjacent driving lane) adjacent to the lane in which the own vehicle is traveling (own vehicle traveling lane). It predicts that you will change lanes to In particular, the behavior prediction of the present embodiment is based on the possibility that the other vehicle will change lanes from the adjacent lane to the own vehicle's lane before the other vehicle actually starts changing lanes. ) to predict the possibility. In the following description, lanes (own vehicle driving lane, adjacent driving lane) are defined with reference to the own vehicle.

The microcomputer 50 is a general-purpose microcomputer having a CPU (Central Processing Unit), a memory, and an input/output unit. A computer program (vehicle behavior prediction program) for functioning as a vehicle behavior prediction device is installed in the microcomputer. By executing the computer program, the microcomputer functions as a plurality of information processing circuits included in the vehicle behavior prediction device. In this embodiment, an example of realizing a plurality of information processing circuits provided in the vehicle behavior prediction device by software is shown. It is also possible to construct an information processing circuit. Also, a plurality of information processing circuits may be configured by individual hardware.

The microcomputer 50 includes a detection integrating section 2a, an object tracking section 2b, an on-map position estimating section 5, a behavior predicting section 10, and a vehicle control section 20 as a plurality of information processing circuits.

The detection integration unit 2a integrates a plurality of detection results obtained from each of the plurality of object detection sensors included in the object detection device 1, and outputs one detection result for each object. Specifically, from the behavior of the object obtained from each of the object detection sensors, the rational behavior of the object that minimizes the error is calculated by considering the error characteristics of each of the object detection sensors. Specifically, by using a known sensor fusion technique, detection results obtained by a plurality of object detection sensors are comprehensively evaluated to obtain more accurate detection results.

The object tracking unit 2b tracks the object detected by the detection integrating unit 2a. Specifically, the object tracking unit 2b verifies (associates) the identity of the object between different times based on the behavior of the object output at different times, and tracks the object based on the correspondence. .

The in-map position estimation unit 5 estimates the position of the vehicle on the map from the absolute position of the vehicle obtained by the vehicle position estimation device 3 and the map information obtained by the map acquisition device 4 . Specifically, the in-map position estimating unit 5 estimates in which lane the host vehicle is traveling.

The behavior prediction unit 10 predicts the possibility of lane change. The behavior prediction unit 10 includes a lane determination unit 11, a vehicle identification unit 12, an intersection position acquisition unit 13, a factor acquisition unit 14, a timing prediction unit 15, a stop time acquisition unit 16, an effect calculation unit 17, and a difficulty level calculation unit 18 .

The lane determining unit 11 is based on the tracking result of the object obtained by the detection integration unit 2a and the object tracking unit 2b and the position of the own vehicle on the map obtained by the in-map position estimating unit 5 (driving lane). to determine in which driving lane on the map other vehicles around the own vehicle are located.

The vehicle identification unit 12 determines whether the vehicle is in an adjacent driving lane (an example of a first driving lane) based on the tracking result of the object obtained by the detection integration unit 2a and the object tracking unit 2b and the determination result of the lane determination unit 11. to identify the stopped vehicle (second other vehicle) that is stopped at the time. It should be noted that the stopped state does not mean only a completely stopped state, but also includes a low speed state that can be regarded as stopped.

In addition, the vehicle identification unit 12 identifies the other vehicle (first other vehicle) for which the possibility of changing lanes is predicted (hereinafter referred to as the "prediction target vehicle") from among the other vehicles traveling in the adjacent lanes. The vehicle identification unit 12 identifies the prediction target vehicle based on the object tracking results obtained by the detection integration unit 2a and the object tracking unit 2b, the determination result of the lane determination unit 11, and the stopping vehicle identification result. conduct.

The intersection position acquisition unit 13 acquires information on the intersection position where the path of the moving object intersects with the adjacent driving lane (road including the adjacent driving lane) based on map information or the like. The intersection position acquisition unit 13 determines whether an intersection position exists in front of the stopped vehicle based on the acquired information. Note that the prediction target vehicle is positioned behind the stopped vehicle, so that the intersection position exists in front of the stopped vehicle is synonymous with the presence of the intersection position in front of the prediction target vehicle.

The factor acquisition unit 14 acquires a stop factor, that is, a factor that the stopped vehicle identified by the vehicle identification unit 12 is stopped based on map information, traffic conditions, and the like. Stopping factors include the passing of an oncoming vehicle whose course intersects the path of the stopped vehicle, the presence of pedestrians in the path of the stopped vehicle, and the occurrence of traffic congestion on the path of the vehicle. mentioned.

The timing prediction unit 15 predicts the timing at which the stopped vehicle can start based on the stop factor acquired by the factor acquisition unit 14 . The timing prediction unit 15 predicts the timing at which the stop factor is eliminated, and based on the predicted timing, predicts the predicted start time, which is the time until the stopped vehicle starts moving.

The stop time acquisition unit 16 acquires the stop time. The stop time refers to the time from identifying the stopped vehicle in the adjacent lane to identifying the prediction target vehicle in the adjacent lane.

The effect calculation unit 17 calculates the effect of lane change. The effect of lane change is the result that the prediction target vehicle can travel in a shorter time by changing lanes (an example of the second lane) than by maintaining the adjacent lane. It is an index that shows

The difficulty calculation unit 18 calculates the difficulty of lane change. The difficulty level of lane change is an index indicating how difficult it is to change lanes when the prediction target vehicle changes lanes from the adjacent lane to the own vehicle's lane.

The lane change calculation unit 19 predicts the possibility of lane change based on the calculation results of the timing prediction unit 15 , the stop time acquisition unit 16 , the effect calculation unit 17 and the difficulty calculation unit 18 .

The vehicle control unit 20 controls the own vehicle based on the possibility of lane change predicted by the behavior prediction unit 10 . The vehicle control unit 20 controls various actuators (steering actuator, accelerator pedal actuator, brake actuator, etc.) of the host vehicle to perform automatic driving control or driving support control (for example, deceleration control).

The vehicle behavior prediction device may include a communication device (not shown). In this case, the behavior prediction unit 10 uses vehicle-to-vehicle communication or road-to-vehicle communication to transmit information ahead of the vehicle (traffic situation information, traffic signal including road information, etc.). Also, the behavior prediction unit 10 may acquire information ahead of the vehicle using information obtained from a camera or the like. In addition, the behavior prediction unit 10 has a database of vehicle behavior data representing average behavior of the vehicle in various driving scenes.

In this embodiment, the microcomputer 50 has the function of the vehicle control unit 20, and the vehicle behavior prediction device can be applied as a vehicle control device. However, the vehicle behavior prediction device may not have the function of the vehicle control unit 20 and may have only the function of predicting the behavior of other vehicles.

Next, a processing procedure for vehicle behavior prediction according to the present embodiment will be described with reference to FIGS. 2 to 10C. This processing procedure is triggered by turning on the ignition switch (IGN) and executed by the microcomputer 50 . In the following description of the processing procedure, the symbols L1, L2, Va, Vb, and Vc shown in the drawings correspond to the own vehicle traveling lane L1, the adjacent traveling lane L2, the own vehicle Va, the prediction target vehicle Vb, and the stopped vehicle Vc. are doing. Also, when identifying a plurality of vehicles, they are described as Vc1, Vc2, and the like.

First, in step S<b>10 , the detection integration unit 2 a acquires object information, that is, information on objects around the vehicle from the object detection device 1 . When the object information is acquired, the detection integration unit 2a calculates the behavior of the object based on the object information. Also, the object tracking unit 2b tracks the object detected by the detection integration unit 2a.

In step S<b>11 , the in-map position estimation unit 5 acquires map information from the map acquisition device 4 . The in-map position estimation unit 5 acquires the position information of the own vehicle Va from the own vehicle position estimation device 3 . When these pieces of information are acquired, the in-map position estimation unit 5 estimates the position (driving lane) of the own vehicle Va on the map.

In step S<b>12 , the lane determination unit 11 acquires the position information of the own vehicle Va estimated by the in-map position estimation unit 5 . Based on the tracking result of the object by the object tracking unit 2b and the position information of the own vehicle Va, the lane determination unit 11 determines to which driving lane on the map the objects around the own vehicle, especially other vehicles, belong. do.

In step S13, the vehicle identification unit 12 identifies the stopped vehicle Vc from other vehicles around the own vehicle Va. Specifically, the vehicle identification unit 12 detects other vehicles whose speed is equal to or less than a speed determination value that is considered to be in a stopped state (for example, a speed determination value such as 1 km/h that can be used to determine that the vehicle is in a substantially stopped state). It is specified as the stopped vehicle Vc. The speed determination value is determined in consideration of the type of sensor provided in the object detection device 1 and detection error. The identification of the stopped vehicle Vc performed by the vehicle identification unit 12 is performed for other vehicles on the adjacent driving lane L2.

Further, when the vehicle identification unit 12 identifies the stopped vehicle Vc, the vehicle identification unit 12 performs a process of identifying the prediction target vehicle Vb. The vehicle identification unit 12 identifies, among the other vehicles in the adjacent driving lane L2, the other vehicle that is expected to affect the behavior of the own vehicle Va due to the lane change to the own vehicle traveling lane L1 as the prediction target vehicle Vb. be done. For example, the vehicle identification unit 12 identifies, as the prediction target vehicle Vb, a vehicle that satisfies the following three conditions among other vehicles traveling in the adjacent driving lane L2. Condition 1 is that the vehicle is running ahead of the own vehicle Va. Condition 2 is that there is a stopped vehicle Vc ahead. Condition 3 is that the inter-vehicle distance to the stopped vehicle Vc is shorter than the assumed distance at which the driver will have the intention of changing lanes.

In step S14, the crossing position acquisition unit 13 acquires the crossing position based on map information or the like. Then, the intersection position acquisition unit 13 determines whether or not there is an intersection position ahead of the stopped vehicle Vc.

The intersection position will be described with reference to FIGS. 3A to 3C. As shown in FIG. 3A, one example of the crossing position is the crossing position where the intersecting road R2 intersects the adjacent traveling lane L2. In this case, the intersecting road R2 corresponds to the route of the moving object intersecting with the adjacent driving lane L2 (the road R1 including this). Further, as shown in FIG. 3B, another example of the crossing position is a position where a side road R4 that can be entered from the adjacent driving lane L2 intersects the adjacent driving lane L2. In this case, the side road R4 corresponds to the path of the moving object that intersects the adjacent driving lane L2 (the road R1 including this). As shown in FIG. 3C, another example of the crossing position is a position adjacent to private land Pa that can be entered from the adjacent driving lane L2 (road R1 including this). The private land Pa includes a facility Pa1 and its parking lot Pa2, etc., and has a path (not shown) of a moving object that moves within the private land Pa. In this case, the private land Pa corresponds to the route of the moving object intersecting the adjacent driving lane L2 (the road R1 including this). In addition to this, another example of the crossing position is a position where pedestrian crossings crossing adjacent traffic lanes intersect. In this case, the pedestrian crossing corresponds to the path of the moving object crossing the adjacent traffic lane.

In step S15, the factor acquiring unit 14 determines whether or not it is difficult to acquire the traffic conditions in front of the stopped vehicle Vc. If the traffic conditions can be acquired, a negative determination is made in step S15, and the process proceeds to step S16. On the other hand, if it is difficult to obtain the traffic conditions, an affirmative determination is made in step S15, and the process proceeds to step S17.

In step S<b>16 , the timing prediction unit 15 predicts the predicted start time based on the traffic conditions and the like acquired by the factor acquisition unit 14 . A processing procedure for predicting the predicted start time will be described with reference to FIGS. 4 and 5. FIG. For example, referring to FIG. 5, consider a scene in which the own vehicle Va passes through an intersection. In FIG. 5, there are three stopped vehicles Vc1, Vc2, and Vc3 in front of the prediction target vehicle Vb.

In step S30, the factor obtaining unit 14 determines whether or not an intersection with a traffic light is ahead of the stopped vehicle Vc based on the map information. If the stop vehicle ahead is an intersection with a traffic light, an affirmative determination is made in step S30, and the process proceeds to step S31. On the other hand, if there is no intersection with a traffic light ahead of the stopped vehicle Vc, a negative determination is made in step S30, and the process proceeds to step S32.

In step S32, the factor acquiring unit 14 acquires the traffic signal state (an example of the traffic situation) at the intersection. The factor acquiring unit 14 acquires the current state of the traffic light, the switching time of the lighting state of the traffic light, and the like.

In step S32, the factor obtaining unit 14 determines whether the stopped vehicle Vc has an intention to turn. Specifically, the factor acquisition unit 14 determines the intention to turn based on the state of the turn signals of the stopped vehicle Vc and whether or not the vehicle is moving closer to the side on the adjacent lane L2. The blinker state of the stopped vehicle Vc can be detected using a camera that captures images of the surroundings of the own vehicle Va, vehicle-to-vehicle communication, or road-to-vehicle communication. The presence or absence of the narrowing can be determined from the position of the stopped vehicle Vc on the adjacent traffic lane L2. When there is an intention to turn, an affirmative determination is made in step S32, and the process proceeds to step S33. On the other hand, if there is no intention to turn, a negative determination is made in step S32, and the process proceeds to step S36.

In step S33, the factor acquisition unit 14 acquires the traffic congestion situation (an example of the traffic situation) where the stopped vehicle Vc turns. The congestion status is information such as the number of congested vehicles stopped at the turn destination, the vehicle speed of the congested vehicles, and the like. The factor acquiring unit 14 may acquire the traffic congestion status using vehicle-to-vehicle communication or road-to-vehicle communication, or may acquire it from information obtained by object detection.

In step S34, the factor acquisition unit 14 acquires information on pedestrians (an example of traffic conditions) existing ahead of the turn of the stopped vehicle Vc. Pedestrian information is information that indicates the situation of pedestrians, such as the position, speed, and number of pedestrians. The factor acquisition unit 14 may acquire pedestrian information using vehicle-to-vehicle communication or road-to-vehicle communication, or may acquire pedestrian information from information obtained by object detection.

In step S35, the factor acquisition unit 14 acquires oncoming vehicle information (an example of traffic conditions). The oncoming vehicle information is information indicating the situation of oncoming vehicles such as the position, vehicle speed, and number of oncoming vehicles traveling in the oncoming lane. The factor acquisition unit 14 may acquire oncoming vehicle information using information obtained by object detection, or may acquire oncoming vehicle information from vehicle-to-vehicle communication or road-to-vehicle communication.

Note that when acquiring oncoming vehicle information using images, it may be difficult to detect oncoming vehicles Vd1 and Vd2 due to the influence of occlusion. In this case, the factor acquiring unit 14 acquires the traffic signal state at the intersection one ahead of the intersection where the stopped vehicles Vc1, Vc2, and Vc3 stop. The traffic signal states to be acquired are the states of the traffic lights that are displayed for traffic in the straight-ahead direction as viewed from the stopped vehicles Vc1, Vc2, and Vc3. Then, the factor acquiring unit 14 acquires oncoming vehicle information based on the signal state of the next intersection. For example, the factor acquiring unit 14 determines that if the traffic light one block ahead is green, there is a high possibility that an oncoming vehicle passing through the intersection one block ahead will approach. On the other hand, if the traffic light is red, the factor acquiring unit 14 determines that the possibility of the oncoming vehicle approaching is low because the oncoming vehicle is stopped at the intersection one ahead.

In step S36, the factor acquiring unit 14 acquires the traffic congestion situation (an example of the traffic situation) in the adjacent lane L2 ahead.

In step S<b>37 , the timing prediction section 15 calculates the predicted start time based on the information acquired by the factor acquisition section 14 . Specifically, the timing prediction unit 15 calculates the predicted start time based on the traffic light state, traffic congestion, pedestrian information, and oncoming vehicle information.

A method of calculating the predicted start time based on the traffic signal state is, for example, as follows. A timing predicting part 15 predicts the time until the signal changes from red to green if the current traffic signal is red, and predicts that stopped vehicles Vc1, Vc2 and Vc3 will start at the timing when the signal changes to green. The timing prediction unit 15 calculates the predicted start times based on the predicted start timings for the stopped vehicles Vc1, Vc2, and Vc3. It should be noted that the time until a red light is switched to a green light can be calculated based on the information from the infrastructure, for example, when information on the time until the light is switched is transmitted from the infrastructure. Alternatively, the elapsed time from the timing when the light turned red to the present can be received from other stopped vehicles via vehicle-to-vehicle communication, and the received duration can be calculated by subtracting the received duration from the average lighting duration of the red signal. Alternatively, the vehicle is equipped with a traffic light detection device that detects the lighting color of the traffic signal, and the elapsed time from the time when it detects that the traffic light has changed to red while approaching the traffic signal to the present is calculated, and the average red light is calculated. can be calculated by subtracting the elapsed time from the lighting duration of . The method of calculating the time until the signal changes from red to green is not limited to the above, and can be calculated or estimated by applying various known methods.

A method for calculating the predicted start time based on traffic congestion information is, for example, as follows. The timing prediction unit 15 predicts the timing at which the vehicle at the end of the traffic jam starts moving based on the number of traffic jam vehicles and the vehicle speed of each traffic jam vehicle. Then, the timing prediction unit 15 predicts that the stopped vehicles Vc1, Vc2, and Vc3 will start at the timing when the vehicles at the end of the traffic jam start to move, and calculates the predicted start time based on the predicted start timing.

A method of calculating the predicted start time based on pedestrian information is, for example, as follows. The timing prediction unit 15 predicts the time required for the pedestrian to cross the road based on the position, speed, number of pedestrians, and width of the road that the pedestrian crosses. The timing prediction unit 15 predicts that the stopped vehicles Vc1, Vc2, and Vc3 will start at the timing when the pedestrian has completely crossed the road, and calculates the predicted start time based on the predicted start timing.

A method of calculating the predicted start time based on the oncoming vehicle information is, for example, as follows. The timing prediction unit 15 predicts the time until the oncoming vehicles Vd1 and Vd2 pass through the intersection based on the positions, speeds and the number of the oncoming vehicles Vd1 and Vd2. The timing prediction unit 15 predicts that the stopped vehicles Vc1, Vc2, and Vc3 will start at the timing at which the oncoming vehicles Vd1 and Vd2 pass, and calculates the predicted start time based on the predicted start timing.

It should be noted that when calculating the start time in consideration of a plurality of situations, the timing prediction unit 15 calculates the predicted start time for each situation, and then determines the longest time as the final predicted start time. good too. Alternatively, the timing prediction unit 15 may determine the final predicted start time by considering a weighting factor for each predicted start time calculated for each situation.

In step S17, the stop time acquisition unit 16 acquires the stop time. After identifying the stopped vehicles Vc1, Vc2, and Vc3 in step S13, the vehicle identification unit 12 identifies the prediction target vehicle Vb when the above conditions are satisfied for the stopped vehicles Vc1, Vc2, and Vc3. The stop time acquisition unit 16 acquires the time from identifying the stopped vehicles Vc1, Vc2, and Vc3 in the adjacent lane L2 until the prediction target vehicle Vb is identified in the adjacent lane L2 as the stop time.

In step S18, the lane change calculator 19 predicts the possibility of lane change. A processing procedure for predicting the possibility of a lane change will be described with reference to FIG.

In step S40, the lane change calculation unit 19 determines whether or not the process of step S16 has been executed and the predicted start time has been predicted. If the predicted start time can be predicted, that is, if the traffic conditions ahead of the stopped vehicles Vc1, Vc2, and Vc3 can be obtained, an affirmative determination is made in step S40, and the process proceeds to step S41. On the other hand, if the predicted start time cannot be predicted, that is, if the traffic conditions ahead of the stopped vehicles Vc1, Vc2, and Vc3 cannot be obtained, a negative determination is made in step S40, and the process proceeds to step S46.

In step S41, the lane change calculation unit 19 determines whether or not the predicted start time is equal to or longer than the first predetermined value. This first predetermined value determines whether the prediction target vehicle Vb is within the time period during which it is considered that the lane change will be postponed even if the stopped vehicles Vc1, Vc2, and Vc3 do not start. The first predetermined value is set by statistically processing human psychological conditions to set the time that can be tolerated without changing lanes (for example, 3 seconds).

If the predicted start time is equal to or longer than the first predetermined value, an affirmative determination is made in step S41, and the process proceeds to step S42. On the other hand, if the predicted start time is shorter than the first predetermined value, a negative determination is made in step S41, and the process proceeds to step S43.

In step S42, the lane change calculation unit 19 calculates the possibility of lane change. The lane change calculation unit 19 has correlation data between the possibility of lane change and the predicted start time, and calculates the possibility of lane change from the predicted start time based on this correlation data.

FIG. 7A illustrates correlation data between the possibility of lane change and the predicted start time. In this correlation data, when the predicted start time is lower than the first predetermined value, the possibility that the prediction target vehicle Vb will change lanes and the possibility that the prediction target vehicle Vb will go straight without changing lanes are the same (0 .5 (50%)). On the other hand, when the predicted start time is equal to or longer than the first predetermined value, it may take time for the stopped vehicles Vc1, Vc2, and Vc3 to start. Therefore, it is set such that the longer the predicted start time, the higher the possibility that the prediction target vehicle Vb will change lanes. According to such correlation data relationship, the lane change calculation unit 19 predicts that the longer the predicted start time, the higher the possibility that the prediction target vehicle Vb will change lanes.

In step S43, the lane change calculation unit 19 acquires the stop time acquired in step S17.

In step S44, the lane change calculation unit 19 calculates the sum of the stop time and the predicted start time (hereinafter referred to as "additional time").

In step S45, the lane change calculation unit 19 calculates the possibility of lane change. The lane change calculation unit 19 has correlation data between the possibility of lane change and additional time, and calculates the possibility of lane change from the additional time based on this correlation data.

FIG. 7B illustrates correlation data between lane change possibility and added time. In this correlation data, when the additional time is shorter than a second predetermined value (for example, 5 seconds), there is a possibility that the prediction target vehicle Vb will change lanes and a possibility that the prediction target vehicle Vb will go straight without changing lanes. Each becomes the same (0.5 (50%)). On the other hand, if the added time is equal to or greater than the second predetermined value, it may take time for the stopped vehicles Vc1, Vc2, and Vc3 to start. Therefore, the longer the addition time, the higher the possibility that the prediction target vehicle Vb will change lanes. According to such correlation data relationship, when the additional time is equal to or greater than the second predetermined value, the lane change calculation unit 19 predicts that the longer the additional time, the higher the possibility that the prediction target vehicle Vb will change lanes. .

The reason for considering the additional time in predicting the possibility of lane change is as follows. Even if the predicted start time is shorter than the first predetermined value, if the driver recognizes the stopped vehicles Vc1, Vc2 and Vc3 for a long time, it is considered that the driver will have the intention to change lanes. Therefore, based on the stop time, the time during which the driver recognizes the stopped vehicles Vc1, Vc2, Vc3 is predicted. By obtaining the sum of the stop time and the predicted start time, the possibility of lane change can be predicted in consideration of the time during which the driver recognizes the stopped vehicles Vc1, Vc2, and Vc3.

In step S46, the lane change calculation unit 19 acquires the stop time acquired in step S17.

In step S47, the lane change calculation unit 19 calculates the possibility of lane change. The lane change calculator 19 has correlation data between the possibility of lane change and stop time, and calculates the possibility of lane change from the stop time based on this correlation data.

FIG. 7C illustrates correlation data between lane change probability and stop time. In this correlation data, when the stop time is shorter than a third predetermined value (for example, 6 seconds), there is a possibility that the prediction target vehicle Vb will change lanes and a possibility that the prediction target vehicle Vb will go straight without changing lanes. Each becomes the same (0.5 (50%)). On the other hand, if the stop time is equal to or longer than the third predetermined value, it may take time for the stopped vehicles Vc1, Vc2, and Vc3 to start. Therefore, the longer the stop time, the higher the possibility that the prediction target vehicle Vb will change lanes. According to such correlation data relationship, when the stop time is equal to or longer than the third predetermined value, the lane change calculation unit 19 predicts that the longer the stop time, the higher the possibility that the prediction target vehicle Vb will change lanes.

In step S19, the effect calculation unit 17 calculates the effect of lane change. Details of the calculation process regarding the effect of lane change will be described below with reference to FIG. 8 .

In step S50, the effect calculation unit 17 sets a reference position. The reference position is a position for calculating the effect of lane change, and is set ahead of the prediction target vehicle Vb. For example, as shown in FIG. 9A, when there is an intersection ahead of the prediction target vehicle Vb, the effect calculation unit 17 sets the intersection position with the intersecting road R2 as the reference position PS. Further, as shown in FIG. 9B, when a side road is connected in front of the prediction target vehicle Vb, the effect calculation unit 17 sets the intersection position with the side road R4 as the reference position PS.

In step S51, the effect calculation unit 17 acquires the state of a traffic signal when there is a traffic signal ahead of the prediction target vehicle Vb. When the traffic light ahead is red, the driver cannot move forward even if he or she changes lanes, and the driver has a mentality of not changing lanes. For example, the effect calculation unit 17 acquires information such as the current lighting color of the traffic light and the signal switching timing. Then, the lane change calculation unit 19 uses these pieces of information to calculate the time required for the prediction target vehicle Vb to pass the reference position PS from the current position, taking into account the state of the traffic light ahead.

In step S52, the effect calculation unit 17 acquires the traffic congestion situation ahead of the prediction target vehicle Vb. When the traffic lane to which the driver is to change lanes is congested, the driver cannot move forward even if the lane is changed, and the psychology of not changing lanes acts on the driver. For example, the lane change calculation unit 19 acquires information such as the length of the traffic jam ahead, the average speed of the traffic jam, and the severity of the traffic jam. Then, the effect calculation unit 17 calculates the time required for the prediction target vehicle Vb to pass through the reference position PS from the current position by using these pieces of information, taking into consideration the traffic congestion situation ahead.

In step S53, the effect calculation unit 17 acquires information on a pedestrian crossing the pedestrian crossing in front of the prediction target vehicle Vb. When a pedestrian is crossing the driving lane of the lane change destination, the driver cannot move forward until the pedestrian has crossed, and the psychology of not changing the lane acts on the driver. For example, the effect calculation unit 17 acquires information such as pedestrian crossing speed and the number of pedestrians. Then, the effect calculation unit 17 uses these pieces of information to calculate the time required for the prediction target vehicle to pass the reference position from the current position, taking into account pedestrians ahead.

In step S54, the effect calculation unit 17 determines, based on the time calculated from steps S51 to S53, the amount of time required for the prediction target vehicle Vb to pass through the reference position PS when it changes lanes to the host vehicle traveling lane L1. A transit time T1 is calculated. For example, the effect calculation unit 17 determines the longest time among the times calculated from steps S51 to S53 as the passage time T1. Alternatively, the effect calculation unit 17 determines the transit time T1 by taking into consideration the weighting coefficients of the times calculated from steps S51 to S53.

In step S55, the effect calculation unit 17 acquires the predicted start time predicted in step S16.

In step S56, the effect calculation unit 17 calculates the passage time T2 required for the target vehicle Vb to pass through the reference position PS when the target vehicle Vb does not change lanes and travels straight through the adjacent lane L2. For example, after the stopped vehicles Vc1, Vc2, and Vc3 have started, the effect calculation unit 17 passes through the position where the stopped vehicles Vc1, Vc2, and Vc3 existed and reaches the reference position PS based on the start prediction time. The transit time T2 up to is calculated.

In step S57, the effect calculation unit 17 calculates the difference ΔT in transit time, specifically, the difference ΔT obtained by subtracting the transit time T2 calculated in step S56 from the transit time T1 calculated in step S54. If the vehicle can pass through the reference position PS in a shorter time if the current driving lane is maintained, the difference ΔT becomes a positive value. On the other hand, if the vehicle can pass through the reference position PS in a shorter time by changing lanes to the own vehicle driving lane L1, the difference ΔT becomes a negative value. That is, it can be said that the larger the difference ΔT in the negative direction, the higher the lane change effect.

In step S20, the difficulty calculation unit 18 calculates the difficulty of changing lanes. In the situation shown in FIG. 10A, there is another vehicle traveling in the host vehicle traveling lane L1 diagonally behind the prediction target vehicle Vb traveling in the adjacent traveling lane L2. In this case, even if the prediction target vehicle Vb changes lanes, it is difficult to secure a sufficient inter-vehicle distance to the vehicle diagonally behind. Therefore, it can be said that the degree of difficulty of changing lanes is high. In addition, in the situation shown in FIG. 10B, the inter-vehicle distance between the last stopped vehicle Vc1 and the prediction target vehicle Vb in the adjacent lane L2 is small. In this case, it is difficult for the prediction target vehicle Vb to change lanes. Therefore, it can be said that the degree of difficulty of changing lanes is high. Further, in the situation shown in FIG. 10C, another vehicle Ve traveling in the host vehicle traveling lane L1 exists diagonally in front of the prediction target vehicle Vb traveling in the adjacent traveling lane L2. In this case, even if the prediction target vehicle Vb changes lanes, it is difficult to secure a sufficient inter-vehicle distance to the preceding vehicle Ve. Therefore, it can be said that the degree of difficulty of changing lanes is high. Therefore, the difficulty level calculation unit 18 calculates the inter-vehicle distance between the other vehicle traveling in the own vehicle traveling lane L1 and the prediction target vehicle Vb traveling in the adjacent traveling lane L2, the stopped vehicle Vc1 in the adjacent traveling lane L2 and the prediction target vehicle Vb. distance between vehicles, etc. Then, the difficulty level calculation unit 18 calculates that the shorter the inter-vehicle distance, the higher the difficulty level of lane change.

In step S21, the lane change calculation unit 19 corrects the possibility of lane change predicted in step S18 based on the calculation results of steps S19 and S20. Specifically, the lane change calculation unit 19 corrects the possibility of lane change so that the lower the effect of lane change, the lower the possibility of lane change. In addition, the lane change calculation unit 19 corrects the possibility of lane change so that the higher the difficulty of lane change, the lower the possibility of lane change.

In step S22, the vehicle control unit 20 controls the own vehicle based on the predicted possibility of lane change. For example, the vehicle control unit 20 performs deceleration control when the possibility of lane change reaches a certain value, for example, 0.8 or more.

In step S23, the lane change calculation unit 19 determines whether or not the ignition switch has been turned off. If the ignition switch is turned off, an affirmative determination is made in step S23, and the series of processing ends (END). On the other hand, if the ignition switch (IGN) is on, a negative determination is made in step S23, and the process returns to step S10.

In the above description, the embodiment has been described by taking as an example a situation in which an intersection exists in front of stopped vehicles Vc1, Vc2, and Vc3. However, as shown in FIG. 11A, the same is true even if there is a side road R4 in the adjacent lane L2 in front of the stopped vehicle Vc. Further, as shown in FIG. 11B, the same applies to the case where private land Pa including facility Pa1 and parking lot Pa2 is adjacent to adjacent lane L2 in front of stopped vehicle Vc. Moreover, the case where there is no such crossing position is also acceptable. For example, as shown in FIG. 11C, there may be stopped vehicles Vc1, Vc2, and Vc3 due to congestion in the adjacent lane L2.

Thus, the vehicle behavior prediction method and vehicle behavior prediction device according to the present embodiment specify the prediction target vehicle Vb for which the possibility of changing lanes is predicted. In addition, the vehicle behavior prediction method and the vehicle behavior prediction device predict the start prediction time, which is the time until the stopped vehicle Vc starts moving, for the stopped vehicle Vc that stops ahead of the prediction target vehicle Vb on the adjacent traveling lane L2. . Then, the vehicle behavior prediction method and the vehicle behavior prediction device predict that the longer the start prediction time is, the higher the possibility that the prediction target vehicle Vb will change lanes to the adjacent driving lane L2.

The driver of the prediction target vehicle Vb initially has a mentality to change lanes when the time until the stopped vehicle Vc starts increases, and then actually starts the lane change operation. Therefore, by predicting the predicted start time, it is possible to predict the possibility of lane change before the lane change operation actually starts. As a result, the lane change of the prediction target vehicle Vb can be predicted early.

Further, in the method of the present embodiment, when the prediction target vehicle Vb changes lanes to the own vehicle traveling lane L1, the deceleration control of the own vehicle is performed based on the predicted possibility of lane change for the prediction target vehicle Vb. Is going.

According to this method, the lane change of the prediction target vehicle Vb can be predicted at an early stage, so the vehicle can be smoothly decelerated without a large change in behavior.

Further, the method of the present embodiment acquires the intersection position where the path of the moving object intersects with the adjacent lane L2, and when the intersection position exists in front of the stopped vehicle Vc (prediction target vehicle Vb), the start prediction is performed. It may be predicted that the longer the time, the higher the possibility of changing lanes.

In this case, the crossing position is the position of the intersection where the intersecting road R2 intersects, the position where the side road R4 that can be entered from the adjacent driving lane L2 intersects, the position adjacent to the private land Pa that can be entered from the adjacent driving lane L2, or the adjacent driving This is the position where the crosswalk crossing the lane L2 intersects.

At the crossing position, a stopped vehicle Vc is likely to occur, and the prediction target vehicle Vb is highly likely to change lanes. By predicting the predicted start time at such an intersection position, the lane change of the prediction target vehicle Vb can be predicted early.

Further, the method of the present embodiment may predict the predicted start time based on the traffic conditions ahead of the stopped vehicle Vc.

Since the stopped vehicle Vc is considered to be stopped under the influence of the traffic in front, the predicted start time can be accurately predicted from the traffic conditions. This makes it possible to appropriately predict the possibility of the prediction target vehicle Vb changing lanes.

In this embodiment, the traffic conditions in front of the stopped vehicle Vc are the conditions of pedestrians crossing the pedestrian crossing on the path of the stopped vehicle Vc, the conditions of the oncoming vehicle Vd traveling in the oncoming lane, and the conditions in front of the stopped vehicle Vc. and a traffic jam situation occurring in front of the stopped vehicle Vc.

According to this method, it is possible to accurately predict the predicted start time, and therefore it is possible to accurately predict the possibility of a lane change.

Further, the method of the present embodiment determines whether or not the traffic conditions in front of the stopped vehicle Vc can be acquired. Then, when the traffic conditions can be acquired, the predicted departure time is predicted based on the traffic conditions. When the traffic condition cannot be acquired, the stop time from the identification of the stopped vehicle Vc stopped in the adjacent lane L2 to the identification of the prediction target vehicle Vb in the adjacent lane L2 is calculated, and based on the stop time. The likelihood of lane changes may be predicted.

If the driver recognizes the stopped vehicle Vc for a long period of time, it is expected that the driver will have the mentality to change lanes. Therefore, by estimating the driver's intention to change lanes according to the stop time, it is possible to determine the possibility of changing lanes before the lane change operation actually starts. Thereby, it is possible to predict the lane change of the prediction target vehicle Vb at an early stage.

Further, the method of the present embodiment may predict that the longer the stop time, the higher the possibility of changing lanes.

When the driver of the prediction target vehicle Vb recognizes that the stopped vehicle Vc is stopped for a long period of time, he/she tends to have a mentality of trying to change lanes. Therefore, by estimating the driver's intention to change lanes according to the stop time, it is possible to determine the possibility of changing lanes before the lane change operation actually starts. As a result, the possibility of lane change can be predicted early.

Further, the method of the present embodiment determines whether or not the predicted start time is equal to or longer than a predetermined value, and if the predicted start time is equal to or longer than the predetermined value, the longer the predicted start time, the more likely it is that the lane can be changed. predicted to be highly likely. On the other hand, the method of the present embodiment may predict the possibility of lane change based on the sum of the stop time and the predicted start time when the predicted start time is shorter than the determination time.

Even if the predicted start time is short, if the time during which the second other vehicle is stopped becomes longer, the driver will tend to change lanes. Therefore, by estimating the driver's intention to change lanes according to the stop time, it is possible to determine the possibility of changing lanes before the lane change operation actually starts. Thereby, it is possible to predict the lane change of the first other vehicle at an early stage.

Further, in the method of the present embodiment, the time required for the prediction target vehicle Vb to pass through the crossing position after changing lanes to the host vehicle traveling lane L1 is predicted as the first passing time T1. In addition, the time required for the prediction target vehicle Vb to pass through the intersection without changing lanes is predicted as a second passing time T2. In this case, the lane change possibility may be modified according to the difference ΔT between the first transit time T1 and the second transit time T2.

If the first transit time T1 is longer than the second transit time T2, the driver may continue to drive in the adjacent lane L2 without changing lanes. Therefore, by correcting the possibility of a lane change according to the time difference, it is possible to make an accurate prediction.

Further, the method of the present embodiment may predict the first transit time T1 based on one or both of the traffic congestion situation in the host vehicle traveling lane L1 and the traffic signal status in front of the stopped vehicle Vc.

The passage time T1 for the vehicle to pass through the lane L1 depends on the traffic congestion in the lane L1 and the traffic signal in front of the vehicle. By considering these situations, the first transit time T1 can be properly predicted. As a result, it is possible to accurately correct the possibility of a lane change.

In addition, the method of the present embodiment may calculate a degree of difficulty indicating the degree of difficulty in changing lanes to the own vehicle driving lane L1, and modify the possibility of changing lanes according to the degree of difficulty.

In a situation in which it is difficult to change lanes to the own vehicle's traveling lane L1, the driver may continue to travel in the adjacent traveling lane L2 without changing lanes. Therefore, by correcting the possibility of lane change according to the degree of difficulty, prediction can be performed with higher accuracy.

In addition, the method of the present embodiment determines the inter-vehicle distance between the prediction target vehicle Vb and another vehicle that is diagonally behind the prediction target vehicle Vb in the host vehicle travel lane L1, The degree of difficulty may be calculated based on the inter-vehicle distance between the other vehicle and the prediction target vehicle Vb, or the inter-vehicle distance between the stopped vehicle Vc and the prediction target vehicle Vb.

When the prediction target vehicle Vb is approaching another vehicle on the host vehicle travel lane L1, the space for changing the course of the vehicle is also limited, making it difficult to change lanes. Therefore, by considering the inter-vehicle distance, it is possible to quantitatively evaluate the difficulty of changing lanes. Possibility of lane change can be corrected with high accuracy. Further, when the prediction target vehicle Vb is approaching the stopped vehicle Vc, the space for changing the course of the vehicle is also limited, making it difficult to change lanes. Therefore, by considering the inter-vehicle distance, it is possible to quantitatively evaluate the difficulty of changing lanes. As a result, the possibility of lane changes can be appropriately corrected. At this time, the degree of difficulty may be calculated in consideration of the number of other vehicles in the host vehicle travel lane L1 that run diagonally behind or diagonally in front of the prediction target vehicle Vb.

In the above-described embodiment, the possibility of the lane change is predicted assuming that the prediction target vehicle Vb traveling in the adjacent lane L2 adjacent to the own vehicle traveling lane L1 changes lanes to the own vehicle traveling lane L1. However, in the vehicle behavior prediction method and the vehicle behavior prediction device according to the present embodiment, it is not necessary for the prediction target vehicle and the own vehicle to exist in adjacent driving lanes. It is sufficient to predict the possibility of changing lanes from a driving lane) to an adjacent driving lane (second driving lane).

While embodiments of the present invention have been described above, the discussion and drawings forming part of this disclosure should not be construed as limiting the invention. Various alternative embodiments, implementations and operational techniques will become apparent to those skilled in the art from this disclosure.

1 Object detection device 2a Detection integration unit 2b Object tracking unit 3 Vehicle position estimation device 4 Map acquisition device 5 On-map position estimation unit 10 Behavior prediction unit 11 Lane determination unit 12 Vehicle identification unit 13 Intersection position acquisition unit 14 Factor acquisition unit 15 Timing prediction unit 16 Stop time acquisition unit 17 Effect calculation unit 18 Difficulty calculation unit 19 Lane change calculation unit 20 Vehicle control unit 50 Microcomputer

Claims (14)

A sensor detects objects including other vehicles around the vehicle,
the computer
Based on map information and information obtained by object detection by the sensor, a first other vehicle that predicts a possibility of changing lanes and the first other vehicle run from among the other vehicles. identifying a second other vehicle that stops in front of the first other vehicle on the first travel lane ;
Obtaining traffic conditions in front of the second other vehicle from a communication device that performs vehicle-to-vehicle communication or road-to-vehicle communication or from information obtained by object detection by the sensor;
acquiring a stop factor that is a factor for the second other vehicle to stop based on the map information and the traffic condition;
predicting the timing at which the stop factor is resolved; predicting the predicted start time, which is the time until the second other vehicle starts, based on the timing at which the stop factor is resolved ;
It is predicted that the longer the predicted start time, the higher the possibility that the first other vehicle will change lanes from the first traveling lane to the second traveling lane adjacent to the first traveling lane. Forecast method. Acquiring an intersection position where the path of the moving object intersects with the first travel lane;
2. The vehicle behavior prediction method according to claim 1, wherein when the intersection position exists in front of the first other vehicle, the longer the predicted start time is, the higher the possibility of the lane change is predicted. The intersection position is a position of an intersection where cross roads intersect, a position where side roads that can be entered from the first traveling lane intersect, a position where private land that can be entered from the first traveling lane is adjacent, or the first 3. The vehicle behavior prediction method according to claim 2, wherein the vehicle behavior prediction method is a position at which a crosswalk crossing the driving lane of the vehicle intersects. The process of specifying the first other vehicle and the second other vehicle includes:
identifying the second other vehicle;
After the second other vehicle is identified, another vehicle that satisfies a predetermined condition is selected from among the other vehicles traveling in the first lane, and the first other vehicle is selected when the predetermined condition is satisfied. including processing to identify as
The predetermined condition is
2. The method includes running ahead of the own vehicle, the presence of the second other vehicle ahead, and the distance between the second other vehicle being shorter than a threshold value. 4. The vehicle behavior prediction method according to any one of 1 to 3. The traffic conditions in front of the second other vehicle include the situation of pedestrians crossing a pedestrian crossing in the path of the second other vehicle, the situation of oncoming vehicles traveling in oncoming lanes, and the situation in front of the second other vehicle. 2. The vehicle behavior prediction method according to claim 1 , further comprising at least one of a state of a traffic light existing in the vehicle and a traffic jam situation occurring ahead of the second other vehicle. Determining whether or not the traffic condition ahead of the second other vehicle can be acquired,
If the traffic conditions can be acquired,
predicting the predicted departure time based on the traffic conditions;
If the traffic conditions cannot be acquired,
Acquiring a stop time that is the time from when the second other vehicle stopped in the first traffic lane is identified until when the first other vehicle is identified in the first traffic lane;
5. The vehicle behavior prediction method according to claim 4 , wherein the possibility of changing lanes is predicted based on the stop time. The vehicle behavior prediction method according to claim 6, wherein the longer the stop time is, the higher the possibility of changing lanes is predicted. Determining whether the predicted start time is equal to or greater than a predetermined value,
When the predicted start time is equal to or greater than the predetermined value,
Predicting that the longer the predicted start time, the higher the possibility of changing the lane,
When the predicted start time is shorter than the predetermined value,
Predicting the possibility of the lane change based on the sum of the stop time, which is the time from the detection of the second other vehicle to the determination of the first other vehicle, and the predicted start time. Item 4. Vehicle behavior prediction method according to item 4 . estimating the time until the first other vehicle passes through the intersection position when the first other vehicle changes lanes to the second traveling lane as a first passage time;
estimating the time until the first other vehicle passes through the intersection position when the first other vehicle does not change lanes as a second passage time;
3. The vehicle behavior prediction method according to claim 2, wherein the possibility of changing lanes is corrected according to the time difference between the first passing time and the second passing time. 10. The vehicle behavior prediction method according to claim 9, wherein the first passage time is predicted based on one or both of traffic congestion conditions in the second travel lane and a traffic signal condition ahead of the first other vehicle. calculating a difficulty level indicating the degree of difficulty in changing lanes from the first driving lane to the second driving lane;
11. The vehicle behavior prediction method according to any one of claims 1 to 10, wherein the possibility of changing lanes is modified according to the degree of difficulty. The inter-vehicle distance between the second other vehicle running diagonally behind the first other vehicle and the first other vehicle in the first running lane, and the distance diagonally in front of the first other vehicle The inter-vehicle distance between the other vehicle on the second driving lane and the first other vehicle on the first driving lane, or the first other vehicle on the first driving lane and the second other vehicle on the first driving lane. 12. The vehicle behavior prediction method according to claim 11, wherein said difficulty level is calculated based on the inter-vehicle distance from the vehicle. a sensor that detects objects including other vehicles around the own vehicle;
a communication device that performs vehicle-to-vehicle communication or road-to-vehicle communication;
a map acquisition device for acquiring map information;
a control unit that predicts the behavior of the other vehicle;
The control unit
a first other vehicle that predicts a possibility of changing lanes from among the other vehicles based on the map information of the map acquisition device and information obtained by object detection by the sensor ; identifying a second other vehicle that stops in front of the first other vehicle on a first traffic lane on which the other vehicle travels;
Acquiring traffic conditions in front of the second other vehicle from the communication device or from information obtained by object detection by the sensor;
acquiring a stop factor that is a factor for the second other vehicle to stop based on the map information and the traffic condition;
predicting the timing at which the stop factor is resolved; predicting the predicted start time, which is the time until the second other vehicle starts, based on the timing at which the stop factor is resolved ;
It is predicted that the longer the predicted start time, the higher the possibility that the first other vehicle will change lanes from the first traveling lane to the second traveling lane adjacent to the first traveling lane. prediction device. a sensor that detects objects including other vehicles around the own vehicle;
a communication device that performs vehicle-to-vehicle communication or road-to-vehicle communication;
a map acquisition device for acquiring map information;
a control unit that predicts the behavior of the other vehicle and controls the vehicle based on the behavior of the other vehicle;
The control unit
a first other vehicle that predicts a possibility of changing lanes from among the other vehicles based on the map information of the map acquisition device and information obtained by object detection by the sensor ; identifying a second other vehicle that stops in front of the first other vehicle on a first traffic lane on which the other vehicle travels;
Acquiring traffic conditions in front of the second other vehicle from the communication device or from information obtained by object detection by the sensor;
acquiring a stop factor that is a factor for the second other vehicle to stop based on the map information and the traffic condition;
predicting the timing at which the stop factor is resolved; predicting the predicted start time, which is the time until the second other vehicle starts, based on the timing at which the stop factor is resolved ;
Predicting that the longer the predicted start time, the higher the possibility that the first other vehicle will change lanes from the first traffic lane to the second traffic lane adjacent to the first traffic lane,
A vehicle control device that performs deceleration control of the own vehicle based on the predicted possibility of lane change when the lane in which the own vehicle is traveling is the second lane.

Images