JP2020158007A - Behavior prediction method and behavior prediction device - Google Patents

Abstract

To improve accuracy of predicting change in a road width direction position of another vehicle.SOLUTION: According to a behavior prediction method, a position and a travel direction of another vehicle 102 around an own vehicle 1 are detected (S10), whether a confluence road 106 with a prescribed width or less, which merges with a travel path 101 of the other vehicle 102 at a point within a prescribed distance ahead of the other vehicle 102 in the travel direction, exists is determined (S13, S14), and when it is determined that the confluence road 106 exists, it is predicted that there is possibility that a road width direction position of the other vehicle 102 moves in a direction separating from one path end 104, which faces the confluence road 106, of path ends 104, 105 of the travel path 101 (S15).SELECTED DRAWING: Figure 2

Description

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

Patent Document 1 detects the shape of the road and the vehicle type of another vehicle, and predicts that when a large vehicle is traveling on a curve, the large vehicle may protrude from the oncoming lane and interfere with the own vehicle. It is stated that it should be done.

Japanese Unexamined Patent Publication No. 2011-204125

However, in addition to the scene where a large vehicle travels on a curve, depending on the position in the road width direction of the own vehicle and the other vehicle, the other vehicle may interfere with the own vehicle or the own vehicle and the other vehicle may be excessively close to each other. there is a possibility. Such a situation can be avoided if the change in the position of another vehicle in the road width direction can be predicted in advance.
An object of the present invention is to improve the accuracy of predicting a change in the position of another vehicle in the road width direction.

In the behavior prediction method according to one aspect of the present invention, the position and the traveling direction of another vehicle around the own vehicle are detected, and the vehicle joins the traveling path of the other vehicle at a predetermined distance ahead of the traveling direction of the other vehicle. It is determined whether or not there is a confluence road that is less than the width, and if it is determined that there is a confluence road, the road width of the other vehicle is in the direction away from one of the roadsides facing the confluence road. Predict that the directional position may move.

According to one aspect of the present invention, it is possible to improve the accuracy of predicting a change in the position of another vehicle in the road width direction.

It is a schematic block diagram of the traveling support device of an embodiment. It is explanatory drawing of an example of the behavior prediction method of an embodiment. It is explanatory drawing of the scene where a confluence road merges on the left side of a driving road without a center line. It is explanatory drawing of the scene where a confluence road merges on the right side of a driving road without a center line. It is a block diagram which shows an example of the functional structure of the traveling support device in embodiment. It is a block diagram which shows an example of the functional structure of the mobility calculation part in FIG. It is explanatory drawing of the roadside object or the depression, the obstacle at the entrance of the confluence road, the position in the road width direction of another vehicle, and the lane width of the traveling lane of another vehicle. It is explanatory drawing of the scene where a motorcycle exists behind another vehicle. It is explanatory drawing of the angle formed by the traveling road of another vehicle and the confluence road. It is explanatory drawing of the conductor of another vehicle. It is explanatory drawing of the conductor of the oncoming vehicle of another vehicle. It is a flowchart of an example of the running support method of an embodiment. It is a flowchart of an example of the other vehicle motion prediction routine. It is a flowchart of an example of a mobility calculation routine.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. It should be noted that each drawing is a schematic one and may differ from the actual one. Further, the embodiments of the present invention shown below exemplify devices and methods for embodying the technical idea of the present invention, and the technical idea of the present invention includes the structure, arrangement, etc. of components. Is not specified as the following. The technical idea of the present invention may be modified in various ways within the technical scope specified by the claims stated in the claims.

(Constitution)
See FIG. The own vehicle 1 includes a running support device 10 that supports the running of the own vehicle 1. The driving support by the traveling support device 10 includes automatic driving control for automatically driving the own vehicle 1 without the driver's involvement based on the driving environment around the own vehicle 1, and driving of the own vehicle 1 by the driver. May include driving assistance control to assist.
The driving support control may include driving control such as automatic steering, automatic braking, constant speed driving control, lane keeping control, and merging support control, as well as outputting a message prompting the driver for steering operation and deceleration operation. ..

The traveling support device 10 includes an object sensor 11, a vehicle sensor 12, a positioning device 13, a map database 14, a communication device 15, a controller 16, an actuator 17, and a notification device 18. In the drawing, the map database is referred to as "map DB".
The object sensor 11 is a plurality of different types that detect an object around the own vehicle 1, such as a laser radar, a millimeter wave radar, a camera, and a LIDAR (Light Detection and Ranging, Laser Imaging Detection and Ranging) mounted on the own vehicle 1. It is equipped with an object detection sensor.

The vehicle sensor 12 is mounted on the own vehicle 1 and detects various information (vehicle signals) obtained from the own vehicle 1. The vehicle sensor 12 includes, for example, a vehicle speed sensor that detects the traveling speed (vehicle speed) of the own vehicle 1, a wheel speed sensor that detects the rotation speed of each tire included in the own vehicle 1, and acceleration in the three axes of the own vehicle 1 ( A 3-axis acceleration sensor (G sensor) that detects deceleration), a steering angle sensor that detects the steering angle (including the steering angle), a gyro sensor that detects the angular speed generated in the own vehicle 1, and a yaw rate that detects the yaw rate. It includes a sensor, an accelerator sensor that detects the accelerator opening of the own vehicle, and a brake sensor that detects the amount of brake operation by the driver.

The positioning device 13 includes a global positioning system (GNSS) receiver, receives radio waves from a plurality of navigation satellites, and measures the current position of the own vehicle 1. The GNSS receiver may be, for example, a Global Positioning System (GPS) receiver or the like. The positioning device 13 may be, for example, an inertial navigation system.
The map database 14 may store high-precision map data (hereinafter, simply referred to as “high-precision map”) suitable as a map for automatic driving. The high-precision map is map data with higher accuracy than the map data for navigation (hereinafter, simply referred to as "navigation map"), and includes lane-based information that is more detailed than road-based information.

For example, in a high-precision map, lane-based information includes lane node information indicating a reference point on a lane reference line (for example, a central line in a lane) and lane link information indicating a lane section mode between lane nodes. including.
The lane node information includes the identification number of the lane node, the position coordinates, the number of connected lane links, and the identification number of the connected lane links. The lane link information includes the lane link identification number, lane type, lane width, lane boundary type, lane shape, lane dividing line shape, and lane reference line shape. High-precision maps also include types and position coordinates of features such as traffic lights, stop lines, signs, buildings, utility poles, curbs, and pedestrian crossings that exist on or near the lane, and lane nodes that correspond to the position coordinates of the features. Includes feature information such as identification numbers for lane links and identification numbers for lane links.

Since the high-precision map includes node and link information for each lane, it is possible to identify the lane in which the own vehicle 1 travels on the traveling route. The high-precision map has coordinates that can represent positions in the extending direction and the width direction of the lane. The high-precision map has coordinates (for example, longitude, latitude and altitude) capable of expressing a position in three-dimensional space, and a lane or the above-mentioned feature may be described as a shape in three-dimensional space.
The communication device 15 performs wireless communication with an external communication device of the own vehicle 1. The communication method by the communication device 15 may be, for example, wireless communication by a public mobile phone network, vehicle-to-vehicle communication, road-to-vehicle communication, or satellite communication.

The controller 16 is an electronic control unit (ECU) that controls the traveling support of the own vehicle 1. The controller 16 includes a processor 21 and peripheral components such as a storage device 22. The processor 21 may be, for example, a CPU (Central Processing Unit) or an MPU (Micro-Processing Unit).
The storage device 22 may include a semiconductor storage device, a magnetic storage device, an optical storage device, and the like. The storage device 22 may include a memory such as a register, a cache memory, a ROM (Read Only Memory) and a RAM (Random Access Memory) used as the main storage device.

The function of the controller 16 described below is realized, for example, by the processor 21 executing a computer program stored in the storage device 22.
The controller 16 may be formed by dedicated hardware for executing each information processing described below.
For example, the controller 16 may include a functional logic circuit set in a general-purpose semiconductor integrated circuit. For example, the controller 16 may have a programmable logic device (PLD: Programmable Logic Device) such as a field-programmable gate array (FPGA).

The actuator 17 operates the steering wheel, accelerator opening degree, and braking device of the own vehicle in response to the control signal from the controller 16 to generate the vehicle behavior of the own vehicle. The actuator 17 includes a steering actuator, an accelerator opening actuator, and a brake control actuator. The steering actuator controls the steering direction and steering amount of the steering of the own vehicle.

The accelerator opening actuator controls the accelerator opening of the own vehicle. The brake control actuator controls the braking operation of the brake device of the own vehicle.
The notification device 18 is an information output device that outputs information (for example, a message prompting a steering operation, an acceleration operation, and a deceleration operation) presented to the driver by the controller 16. The notification device 18 may include, for example, a display device that outputs visual information, a lamp or a meter, or a speaker that outputs audio information.

Next, an example of traveling support control by the controller 16 will be described. FIG. 2 shows a scene in which another vehicle 102 is traveling on the first traveling lane 101 of the traveling lane 100 and the own vehicle 1 is traveling on the second traveling lane 103 which is an oncoming lane of the first traveling lane 101. ing.
In front of the traveling direction of the other vehicle, a confluence road 106 having a road width Wm joins one of the roadsides 104 and 105 of the traveling road 100, which is closer to the first traveling lane 101, and the other vehicle 102. May turn to the confluence road 106.

Here, if the road width Wm of the merging road 106 is narrow, the driver of the other vehicle 102 turning to the merging road 106 once steers in the direction opposite to the direction of turning to the merging road 106, and then reverses the steering direction. A steering operation such as steering in a turning direction to the confluence road 106 may be performed, and the other vehicle 102 may travel on a track as shown by the one-point chain line 107. Hereinafter, such a steering operation will be referred to as "temporary reversal steering".
When the temporary reverse steering is performed and the position of the other vehicle 102 on the travel path 100 in the road width direction moves, the other vehicle 102 may interfere with the own vehicle 1 or the own vehicle 1 and the other vehicle 102 may be excessively close to each other. ..

In the example of FIG. 2, the other vehicle 102 protrudes into the second traveling lane 103, which is an oncoming lane, and interferes with the own vehicle 1, or the own vehicle 1 and the other vehicle 102 excessively approach the second traveling lane 103. May approach.
Further, for example, when the own vehicle 1 traveling behind the other vehicle 102 overtakes the other vehicle 102, the own vehicle 1 may be hindered or the own vehicle 1 and the other vehicle 102 may be excessively close to each other.

Therefore, the controller 16 determines whether or not there is a confluence road 106 having a predetermined width or less that merges with the travel path 100 of the other vehicle 102 at a point within a predetermined distance ahead of the traveling direction of the other vehicle 102.
When it is determined that such a merging road 106 exists, the controller 16 determines that the road width of the other vehicle 102 is away from one of the roadsides 104 and 105 of the traveling road 100 facing the merging road 106. Predict that the directional position may move.

When it is predicted that the position of the other vehicle 102 in the road width direction may move, the controller 16 determines the target travel of the own vehicle 1 to increase the distance between the own vehicle 1 and the other vehicle 102 in the road width direction of the travel path 100. A track is generated, and travel control is executed to allow the own vehicle 1 to travel along the target travel track. Alternatively, a speed profile for decelerating the own vehicle 1 is generated, and the speed of the own vehicle is controlled according to this speed profile.
As a result, even if the position of the other vehicle 102 in the road width direction moves, sudden steering control and speed control can be avoided, so that the own vehicle 1 can run smoothly.

Although FIG. 2 shows an example of the traveling path 100 having the center line, it is conceivable that the driver of the other vehicle 102 temporarily reverses the steering even on the traveling path without the center line.
In the case of a traveling road without a center line, as shown in FIGS. 3A and 3B, regardless of whether the merging road 106 merges on the left side or the right side of the traveling road 100, the driver of the other vehicle 102 temporarily There is a risk of reverse steering.

Therefore, the controller 16 predicts that the position of the other vehicle 102 in the road width direction may move even if it is determined that there is a confluence road 106 that joins the traveling road without the center line.
For example, in the case of FIG. 3A, as in the case of FIG. 2, the position in the road width direction of the other vehicle 102 in the direction away from one of the roadsides 104 and 105 of the traveling road 100 facing the confluence road 106. Predict that may move.
For example, in the case of FIG. 3B, there is a possibility that the position of the other vehicle 102 in the road width direction may move in a direction away from one of the roadsides 108 and 109 of the traveling road 100 facing the confluence road 106. Predict.

Subsequently, the function of the controller 16 will be described in detail with reference to FIG. The controller 16 includes an object detection unit 30, an own vehicle position estimation unit 31, a map acquisition unit 32, a detection integration unit 33, an object tracking unit 34, an in-map position calculation unit 35, and an operation prediction unit 36. It includes a own vehicle route generation unit 37 and a vehicle control unit 38.
The object detection unit 30 detects the position, posture, size, speed, and the like of objects around the own vehicle 1, such as a vehicle, a motorcycle, a pedestrian, and an obstacle, based on the detection signal of the object sensor 11. The object detection unit 30 outputs a detection result representing a two-dimensional position, posture, size, speed, etc. of an object in, for example, a zenith view (also referred to as a plan view) in which the own vehicle 1 is viewed from the air.

The own vehicle position estimation unit 31 determines the absolute position of the own vehicle 1, that is, the position of the own vehicle 1 with respect to a predetermined reference point, based on the measurement result by the positioning device 13 and the odometry using the detection result from the vehicle sensor 12. , Measure posture and speed.
The map acquisition unit 32 acquires map information indicating the structure of the road on which the own vehicle 1 travels from the map database 14. The map acquisition unit 32 may acquire map information from an external map data server by the communication device 15.

The detection integration unit 33 integrates a plurality of detection results obtained by the object detection unit 30 from each of the plurality of object detection sensors, and outputs one detection result for each object.
Specifically, from the behavior of the object obtained from each of the object detection sensors, the most rational behavior of the object with the least error is calculated in consideration of the error characteristics of each object detection sensor.
Specifically, by using a known sensor fusion technique, the detection results acquired by a plurality of types of sensors are comprehensively evaluated to obtain more accurate detection results.

The object tracking unit 34 tracks the object detected by the object detecting unit 30. Specifically, based on the detection result integrated by the detection integration unit 33, the identity of the object between different times is verified (associated) from the behavior of the objects output at different times, and the same is performed. Based on the association, the behavior such as the velocity of the object is predicted.

The position calculation unit 35 in the map estimates the position and attitude of the own vehicle 1 on the map from the absolute position of the own vehicle 1 obtained by the own vehicle position estimation unit 31 and the map information acquired by the map acquisition unit 32. To do.
In addition, the position calculation unit 35 in the map identifies the road on which the own vehicle 1 is traveling and the lane in which the own vehicle 1 is traveling on the road.

The motion prediction unit 36 is based on the detection results obtained by the detection integration unit 33 and the object tracking unit 34 and the position of the own vehicle 1 specified by the position calculation unit 35 in the map, and another other in the vicinity of the own vehicle 1. Predict the movement of an object. The motion prediction unit 36 is an example of the behavior prediction device described in the claims.
The motion prediction unit 36 includes a lane determination unit 40, a right / left turn intention prediction unit 41, a confluence road detection unit 42, a confluence road width calculation unit 43, a mobility calculation unit 44, and a track prediction unit 45.

The lane determination unit 40 is based on the detection results obtained by the detection integration unit 33 and the object tracking unit 34 and the position of the own vehicle 1 specified by the position calculation unit 35 in the map, and the lane determination unit 40 of the object around the own vehicle 1. Estimate the position and direction of travel on the map. Then, it is determined which lane in the map the object belongs to.
The right / left turn intention prediction unit 41 predicts whether or not the other vehicle 102, which is an object around the own vehicle, is about to turn.

For example, the right / left turn intention prediction unit 41 makes a song based on whether or not the winker of another vehicle 102 is lit by image recognition based on a camera mounted on the own vehicle 1 (for example, a camera included in the object sensor 11). You may determine your intention to try.
Specifically, the right / left turn intention prediction unit 41 may determine whether or not the other vehicle 102 is about to turn based on the lighting of the winker indicating the intention to turn to the confluence road 106 among the left and right winkers. Further, for example, when the center line exists as shown in FIG. 2, the judgment is made based on the lighting of the left winker on the driving road where the left side traffic is obligatory, and the right side on the driving road where the right side traffic is obligatory. The judgment may be made based on the lighting of the winker.

Further, is the right / left turn intention prediction unit 41 a deceleration pattern for turning the speed pattern of the other vehicle 102 based on the speed profile of the other vehicle 102 obtained from the detection results of the detection integration unit 33 and the object tracking unit 34? By determining whether or not, the intention to bend may be determined.
When it is determined that the other vehicle 102 is about to turn, the confluence road detection unit 42 presents the other vehicle 102 based on the position and traveling direction of the other vehicle 102 and the map information acquired by the map acquisition unit 32. It is determined whether or not the merging road 106 exists at a point within a predetermined distance from the other vehicle 102 ahead of the traveling direction of the first traveling lane 101, and the merging road 106 is detected.

When a plurality of confluence road candidates are detected, the blinker lighting timing and deceleration start timing for each confluence road candidate position are predicted and compared with the actual lighting timing and deceleration start timing of the other vehicle 102. , The confluence road 106 may be specified from the candidates for the confluence road.
Alternatively, instead of this, the processing described below may be performed in the same manner for each of the plurality of confluence roads 106.
The confluence road 106 may be a connecting portion with a private road such as the entrance of a parking lot.

The confluence road width calculation unit 43 calculates the road width Wm of the confluence road 106 detected by the confluence road detection unit 42 by using map information, LiDAR, a stereo camera, or the like. For example, the confluence road width calculation unit 43 calculates the road width at the entrance of the confluence road 106 connected to the first traveling lane 101.
The mobility calculation unit 44 calculates the possibility that the position of the other vehicle 102 in the road width direction moves away from one of the roadsides 104 and 105 of the traveling road 100 facing the confluence road 106. ..

See FIG. The mobility calculation unit 44 includes a confluence road width determination unit 50, a roadside condition determination unit 51, an entrance condition determination unit 52, a road width direction position determination unit 53, a traveling lane width determination unit 54, and a two-wheeled vehicle determination unit. It includes 55, an angle determination unit 56, a vehicle length determination unit 57, an oncoming vehicle determination unit 58, and a possibility estimation unit 59.
The confluence road width determination unit 50 determines whether or not the road width Wm of the confluence road 106 is equal to or less than a predetermined width. This is because the narrower the road width Wm of the confluence road 106, the higher the possibility that the driver of the other vehicle 102 will perform temporary reverse steering. The predetermined width may be, for example, a general road width of 3.0 m.

The roadside condition determination unit 51 determines the height of an object existing at the roadside 104 near the entrance of the confluence road 106 and the depth of the depression. If there is something that cannot be driven, such as a curb, a fence, or a gutter, at the roadside 104 near the entrance of the confluence road 106, another vehicle may derail, and in order to avoid this, the driver of the other vehicle 102 This is because the possibility of temporarily reverse steering is increased.

See FIG. The roadside condition determination unit 51 determines whether or not the roadside 104 near the entrance of the confluence road 106 has an object 112 having a height equal to or higher than a predetermined height or a depression 112 having a depth equal to or higher than a predetermined depth. To do.
For example, the roadside condition determination unit 51 can travel near the entrance of the confluence road 106, for example, at a position where there is a possibility of derailing due to an inner ring difference when another vehicle turns, or the height of the object 112 or the depth of the depression 112. Determine if it is height or depth.

The entrance state determination unit 52 is an obstacle at the entrance of the confluence road 106 (for example, the turning center side of another vehicle 102 that turns to the confluence road 106) to narrow the width of the travelable range of the entrance of the confluence road 106, such as a construction cone or a bicycle. It is determined whether or not the object 111 exists.
This is because the narrower the width of the travelable range at the entrance of the confluence road 106, the higher the possibility that the driver of the other vehicle 102 will perform temporary reverse steering.

The road width direction position determination unit 53 determines the road width direction position Pw of another vehicle 102 on the travel path 100. This is because the closer the position Pw in the road width direction of the other vehicle 102 is to the roadside 104, the higher the possibility that the driver of the other vehicle 102 will temporarily reverse the steering.
The road width direction position Pw may be represented by, for example, the road width direction distance from the boundary position between the roadside zone and the roadway to the other vehicle 102, and the road width direction distance from the center of the traveling road 100 or the first traveling lane 101 to the other vehicle 102. It may be represented by.

The traveling lane width determining unit 54 determines whether or not the lane width WL of the first traveling lane 101, which is the traveling lane of the other vehicle 102, is equal to or less than a predetermined value. This is because when the lane width WL of the first traveling lane 101 is narrow, the other vehicle 102 becomes closer to the roadside 104, and the possibility that the driver of the other vehicle 102 temporarily reverse steering is increased.
The motorcycle determination unit 55 determines whether or not there is a motorcycle behind the other vehicle 102. See FIG. 7. When the motorcycle 113 is present behind the other vehicle 102, it often travels closer to the roadside 104 side in order to avoid getting caught when turning to the confluence road 106. Therefore, the other vehicle 102 becomes closer to the roadside 104, and the possibility that the driver of the other vehicle 102 temporarily reverse steering is increased.

The angle determination unit 56 determines whether or not the angle formed by the traveling road 101 and the confluence road 106 is an acute angle.
See FIG. The angle determination unit 56 determines whether or not one of the angles θ1 and θ2 formed by the traveling road 101 and the confluence road 106, which is closer to the other vehicle 102, is an acute angle.
When the angle θ1 formed by the traveling road 101 and the confluence road 106 is an acute angle, the inner ring is likely to derail or collide with an object at the roadside 104 near the entrance of the confluence road 106, and the driver of the other vehicle 102 This is because the possibility of temporarily reverse steering is increased.

See FIG. The vehicle length determination unit 57 determines whether or not the vehicle length L1 of the other vehicle 102 is longer than the predetermined length.
This is because when the length L1 of the other vehicle 102 is long, the inner ring difference becomes large, and the inner ring is likely to derail or collide with an object at the roadside 104 near the entrance of the confluence road 106.
If the length L1 of the other vehicle 102 cannot be measured directly from the own vehicle 1 due to occlusion, whether the length L1 is longer than the predetermined length based on the height and width of the front of the other vehicle 102. May be determined.

See FIG. The oncoming vehicle determination unit 58 determines whether or not the vehicle length L2 of the oncoming vehicle 114 of the other vehicle 102 is longer than the predetermined length. The oncoming vehicle 114 may be a vehicle on the second traveling lane 103 in front of or behind the own vehicle 1.
This is because when there is an oncoming vehicle 114 having a long vehicle length L2, the possibility that the driver of the other vehicle 102 temporarily reverse steering is reduced.

The possibility estimation unit 59 estimates the possibility that the road width direction position Pw of the other vehicle 102 moves away from the roadside 104 facing the confluence road 106, that is, the possibility that the driver makes a temporary reverse steering. ..
When the confluence road width determination unit 50 determines that the road width Wm of the confluence road 106 is equal to or less than a predetermined width, the possibility estimation unit 59 moves the position Pw in the road width direction of the other vehicle 102 in the direction away from the roadside 104. Judge that there is a possibility.

Further, the possibility estimation unit 59 determines whether there is an object 112 having a height equal to or higher than a predetermined height or a depression 112 having a depth equal to or higher than a predetermined depth at the roadside 104 near the entrance of the confluence road 106. When the determination unit 51 determines, it is determined that there is a high possibility that the road width direction position Pw of the other vehicle 102 will move in the direction away from the roadside 104.
Further, when the possibility estimation unit 59 determines that the entrance of the confluence road 106 has an obstacle 111 that narrows the width of the travelable range of the entrance of the confluence road 106, the possibility estimation unit 59 determines that the other vehicle It is determined that there is a high possibility that the road width direction position Pw of 102 moves in the direction away from the roadside 104.

Further, in the possibility estimation unit 59, the closer the road width direction position Pw of the other vehicle 102 determined by the road width direction position determination unit 53 is to the roadside 104, the farther the road width direction position Pw of the other vehicle 102 is from the roadside 104. Judge that there is a high possibility of moving.
Further, when the traveling lane width determining unit 54 determines that the lane width WL of the first traveling lane 101, which is the traveling lane of the other vehicle 102, is equal to or less than a predetermined value, the possibility estimation unit 59 determines in the road width direction of the other vehicle 102. It is determined that there is a high possibility that the position Pw will move away from the roadside 104.

Further, when the motorcycle determination unit 55 determines that a motorcycle exists behind the other vehicle 102, the possibility estimation unit 59 may move the position Pw in the road width direction of the other vehicle 102 in the direction away from the roadside 104. Judged as high.
Further, when the angle determination unit 56 determines that the angle θ1 formed by the traveling road 101 and the confluence road 106 is an acute angle, the possibility estimation unit 59 indicates that the road width direction position Pw of the other vehicle 102 is away from the roadside 104. Judge that there is a high possibility of moving to.

Further, when the vehicle length determination unit 57 determines that the vehicle length L1 of the other vehicle 102 is longer than the predetermined length, the possibility estimation unit 59 moves the position Pw in the road width direction of the other vehicle 102 in the direction away from the roadside 104. Judge that the possibility is high.
Further, when the oncoming vehicle determination unit 58 determines that the vehicle length L2 of the oncoming vehicle 114 of the other vehicle 102 is longer than the predetermined length, the possibility estimation unit 59 sets the road width direction position Pw of the other vehicle 102 from the roadside 104. It is determined that there is a high possibility of moving in the direction of separation.

Based on each of the above determination results, the possibility estimation unit 59 estimates the high possibility that the road width direction position Pw of the other vehicle 102 will move in the direction away from the roadside 104.
For example, the possibility estimation unit 59 may estimate the possibility Pa that the position Pw in the road width direction of the other vehicle 102 moves away from the roadside 104 based on the following equation (1).
Pa = Pr + Ar x Xr + Ai x Xi + Ap x Xp + Aw x Xw + Am x Xm + Aa x Xa + AL1 x XL1-AL2 x XL2 ... (1)

Here, Pr is a basic probability (constant), and Ar, Ai, Ap, Aw, Am, Aa, AL1, and AL2 are positive coefficients.
Further, Xr, Xi, Xp, Xw, Xm, Xa, XL1 and XL2 are the roadside condition determination unit 51, the entrance condition determination unit 52, the road width direction position determination unit 53, and the traveling lane width determination unit 54, respectively. This is a variable indicating the determination results of the two-wheeled vehicle determination unit 55, the angle determination unit 56, the vehicle length determination unit 57, and the oncoming vehicle determination unit 58.

For example, the value of the variable Xr is "1" when there is an object 112 having a height equal to or higher than a predetermined height or a depression 112 having a depth equal to or higher than a predetermined depth at the roadside 104 near the entrance of the confluence road 106. Yes, otherwise it is "0".
For example, the value of the variable Xi is "1" when there is an obstacle 111 that narrows the width of the travelable range at the entrance of the confluence road 106, and is "0" in other cases.

For example, the variable Xp has a value that increases as the road width direction position Pw of the other vehicle 102 approaches the roadside 104.
For example, the value of the variable Xw is "1" when the lane width WL of the first traveling lane 101 is equal to or less than a predetermined value, and is "0" in other cases.
For example, the value of the variable Xm is "1" when there is a motorcycle behind the other vehicle 102, and "0" in other cases.

For example, the value of the variable Xa is "1" when the angle θ1 formed by the traveling road 101 and the confluence road 106 is an acute angle, and is “0” in other cases.
For example, the value of the variable XL1 is "1" when the length L1 of the other vehicle 102 is longer than the predetermined length, and is "0" in other cases.
For example, the value of the variable XL2 is "1" when the length L2 of the oncoming vehicle 114 of the other vehicle 102 is longer than the predetermined length, and is "0" in other cases.

See FIG. The trajectory prediction unit 45 is based on the detection results obtained by the detection integration unit 33 and the object tracking unit 34 and the position of the own vehicle 1 specified by the position calculation unit 35 in the map, and other parts around the own vehicle 1. Predict the running trajectory of the vehicle.
At this time, the track prediction unit 45 determines that the road width direction position Pw of the other vehicle 102 may move in the direction away from the roadside 104, or when the possibility Pa is equal to or more than the threshold value, the other vehicle The traveling track of the other vehicle 102 is predicted when the driver of the 102 temporarily reverses the steering.
For example, the track prediction unit 45 reverses the steering direction after the other vehicle 102 is steered in the direction opposite to the direction of turning to the confluence road 106, the road width direction position Pw is moved away from the roadside 104 by a predetermined length, and then the confluence road. The traveling track generated when the vehicle is steered in the direction of turning to 106 is calculated.

The own vehicle route generation unit 37 uses traffic rules (for example, traveling along the second traveling lane 103, etc.) based on the prediction result of the movement of other objects around the own vehicle 1 predicted by the motion prediction unit 36. While following the above, a smooth target traveling trajectory and speed profile are generated so that the own vehicle 1 can travel without colliding with another object and without sudden deceleration and sudden steering due to the behavior of the other object.
For example, the own vehicle route generation unit 37 increases the distance between the own vehicle 1 and the other vehicle 102 in the road width direction when the traveling track when the driver of the other vehicle 102 temporarily reverses steering is predicted. A target traveling track may be generated. Alternatively, a speed profile for decelerating the own vehicle may be generated.

The vehicle control unit 38 drives the actuator 17 so that the own vehicle 1 travels on the target traveling track at a speed according to the speed profile generated by the own vehicle route generation unit 37.
The travel control of the vehicle control unit 39 does not necessarily require a target travel track and a speed profile. For example, braking control, acceleration control, and steering control based on the relative distance to an object (for example, an obstacle) around the own vehicle 1 are also possible.

The motion prediction unit 36 notifies the notification device 18 when it is determined that the road width direction position Pw of the other vehicle 102 may move in the direction away from the roadside 104, or when the possibility Pa is equal to or greater than the threshold value. May be activated to notify the occupants of the own vehicle 1 of the alarm.
In this case, the notification device 18 may output, for example, a voice message or a visual message informing that the road width direction position Pw of the other vehicle 102 may move away from the roadside 104. A voice message or a visual message prompting a steering operation for increasing the distance between the own vehicle 1 and the other vehicle 102 and a deceleration operation for the own vehicle 1 may be output.

Next, an example of the operation of the traveling support device 10 in the embodiment will be described with reference to FIG.
In step S1, the object detection unit 30 detects the position, posture, size, speed, and the like of the object around the own vehicle 1 by using a plurality of object detection sensors.
In step S2, the detection integration unit 33 integrates a plurality of detection results obtained from each of the plurality of object detection sensors, and outputs one detection result for each object. The object tracking unit 34 tracks each detected and integrated object, and predicts the behavior of the objects around the own vehicle 1.

In step S3, the own vehicle position estimation unit 31 measures the position, posture, and speed of the own vehicle 1 with respect to a predetermined reference point based on the measurement result by the positioning device 13 and the odometry using the detection result from the vehicle sensor 12. To do.
In step S4, the map acquisition unit 32 acquires map information indicating the structure of the road on which the own vehicle 1 travels.
In step S5, the position calculation unit 35 in the map estimates the position and posture of the own vehicle 1 on the map from the position of the own vehicle 1 measured in step S3 and the map data acquired in step S4.

In step S6, the motion prediction unit 36 is based on the detection result (behavior of an object around the own vehicle 1) obtained in step S2 and the position of the own vehicle 1 specified in step S5, and is in the vicinity of the own vehicle 1. Predict the movement of other vehicles in.
The other vehicle motion prediction routine by the motion prediction unit 36 will be described later with reference to FIGS. 12 and 13.
In step S7, the own vehicle route generation unit 37 generates the target traveling track and speed profile of the own vehicle 1 based on the movement of the other vehicle predicted in step S6.
In step S8, the vehicle control unit 38 controls the own vehicle 1 so that the own vehicle 1 travels according to the target traveling track and the speed profile generated in step S7.

The other vehicle motion prediction routine of FIG. 11 will be described with reference to FIG.
In step S10, the lane determination unit 40 is in the vicinity of the own vehicle 1 based on the detection result obtained by the detection integration unit 33 and the object tracking unit 34 and the position of the own vehicle 1 specified by the position calculation unit 35 in the map. The position and traveling direction of the object on the map are determined, and which lane in the map the object belongs to is determined.

In step S11, the right / left turn intention prediction unit 41 determines whether or not among the winkers of the other vehicle 102, which is an object around the own vehicle, the winker indicating the intention to turn to the confluence road 106 is lit. When the blinker is lit (step S11: Y), the process proceeds to step S13. When the blinker is not lit (step S11: N), the process proceeds to step S12.

In step S12, the right / left turn intention prediction unit 41 determines whether or not the speed pattern of the other vehicle 102 is a deceleration pattern for turning. In the case of the deceleration pattern (step S12: Y), the process proceeds to step S13. If there is no deceleration pattern (step S12: N), the process proceeds to step S17.
In step S13, the confluence road detection unit 42 is within a predetermined distance from the other vehicle 102 ahead of the traveling direction of the first traveling lane 101 in which the other vehicle 102 exists, based on the position and the traveling direction of the other vehicle 102 and the map information. Identify the confluence road 106 that exists at the point.

In step S14, the confluence road width calculation unit 43 calculates the road width Wm of the confluence road 106 detected in step S13. The confluence road width determination unit 50 determines whether or not the road width Wm of the confluence road 106 is equal to or less than a predetermined width. When the road width Wm of the confluence road 106 is equal to or less than a predetermined width (step S14: Y), the process proceeds to step S15. When the road width Wm of the confluence road 106 is not equal to or less than the predetermined width (step S14: N), the process proceeds to step S17.

In step S15, the mobility calculation unit 44 determines that the road width direction position Pw of the other vehicle 102 may move in a direction away from one road end 104 facing the confluence road 106, and determines that the road width direction position Pw Calculates the possibility Pa of moving away from the roadside 104.
The mobility calculation routine by the mobility calculation unit 44 will be described with reference to FIG.

In step S16, the track prediction unit 45 determines the periphery of the own vehicle 1 based on the detection results obtained by the detection integration unit 33 and the object tracking unit 34 and the position of the own vehicle 1 specified by the position calculation unit 35 in the map. Predict the trajectory of other vehicles.
At this time, if it is determined that the road width direction position Pw of the other vehicle 102 may move away from the roadside 104, or if the possibility Pa is equal to or greater than the threshold value, the driver of the other vehicle 102 The traveling trajectory of the other vehicle 102 when the temporary reverse steering is performed is predicted.

In step S17, the motion prediction unit 36 determines whether or not the processing of steps S10 to S16 has been performed on all the objects around the own vehicle 1. When steps S10 to S16 are performed for all the objects (S17: Y), the other vehicle motion prediction routine is terminated. When steps S10 to S16 have not been performed for any of the other vehicles (S17: N), the process selects an object that has not been processed as the processing target, and the processing returns to step S10.

The mobility calculation routine of FIG. 12 will be described with reference to FIG.
In step S20, the roadside condition determination unit 51 determines whether or not the roadside 104 near the entrance of the confluence road 106 has an object 112 having a height equal to or higher than a predetermined height or a depression 112 having a depth equal to or higher than a predetermined depth. Is determined.
In step S21, the entrance state determination unit 52 determines whether or not there is an obstacle 111 that narrows the width of the travelable range of the entrance of the confluence road 106 at the entrance of the confluence road 106.

In step S22, the road width direction position determination unit 53 determines the road width direction position Pw of the other vehicle 102 on the travel path 100.
In step S23, the traveling lane width determining unit 54 determines whether or not the lane width WL of the first traveling lane 101, which is the traveling lane of the other vehicle 102, is equal to or less than a predetermined value.
In step S24, the motorcycle determination unit 55 determines whether or not a motorcycle exists behind the other vehicle 102.

In step S25, the angle determination unit 56 determines whether or not the angle θ1 formed by the traveling road 101 and the merging road 106 is an acute angle.
In step S26, the vehicle length determination unit 57 determines whether or not the vehicle length L1 of the other vehicle 102 is longer than the predetermined length.

In step S27, the oncoming vehicle determination unit 58 determines whether or not the vehicle length L2 of the oncoming vehicle 114 of the other vehicle 102 is longer than the predetermined length.
In step S28, the possibility estimation unit 59 calculates the possibility Pa that the road width direction position Pw of the other vehicle 102 moves away from the roadside 104 according to the determination results of steps S20 to S27. After that, the mobility calculation routine ends.

(Effect of embodiment)
(1) The lane determination unit 40 detects the position and the traveling direction of another vehicle 102 around the own vehicle 1. The confluence road detection unit 42, the confluence road width calculation unit 43, and the confluence road width determination unit 50 merge with the travel path 101 of the other vehicle 102 at a point within a predetermined distance ahead of the traveling direction of the other vehicle 102, and have a predetermined width or less. It is determined whether or not the confluence road 106 exists. When it is determined that the merging road 106 exists, the mobility calculation unit 44 determines that the other vehicle 102 moves away from one of the roadsides 104 and 105 of the traveling road 101 facing the merging road 106. Predict that the road width direction position Pw may move,

As a result, before the other vehicle 102 turns to the narrow confluence road 106, the possibility that the road width direction position Pw of the other vehicle 102 temporarily moves in the direction opposite to the turning direction can be predicted in advance, so that the road width of the other vehicle 102 can be predicted. The accuracy of predicting the change of the directional position Pw can be improved.
As a result, smooth running can be realized by avoiding that the other vehicle 102 interferes with the own vehicle 1 or the own vehicle 1 and the other vehicle 102 are excessively close to each other.

(2) When it is determined that the merging road 106 exists and the direction indicator of the other vehicle 102 is lit, the mobility calculation unit 44 determines that the other vehicle 102 moves away from the roadside 104. It is predicted that the position Pw in the road width direction may move.
As a result, the intention of the other vehicle 102 to turn to the confluence road 106 can be estimated, so that the possibility of steering in the direction opposite to the direction of turning to the narrow confluence road 106 can be accurately estimated.

(3) When it is determined that the merging road 106 exists and the other vehicle 102 is decelerating, the mobility calculation unit 44 determines that the position Pw of the other vehicle 102 in the road width direction away from the roadside 104. Predict that may move.
As a result, the intention of the other vehicle 102 to turn to the confluence road 106 can be estimated, so that the possibility of steering in the direction opposite to the direction of turning to the narrow confluence road 106 can be accurately estimated.

(4) Possibility estimation when it is determined that an object having a height equal to or higher than a predetermined height with respect to the traveling road surface or a depression having a depth equal to or higher than a predetermined depth with respect to the traveling road surface exists at the roadside 104. The unit 59 predicts that there is a high possibility that the road width direction position Pw of the other vehicle 102 will move in the direction away from the roadside 104.
As a result, it is possible to accurately estimate the possibility of steering in the direction opposite to the direction of turning to the confluence road 106 in an attempt to avoid an object or a dent at the roadside 104 at the entrance of the confluence road 106.

(5) When there is an obstacle at the entrance of the confluence road 106 that narrows the travelable range in the width direction of the confluence road, the possibility estimation unit 59 moves away from the roadside 104 in the road width direction of the other vehicle 102. It is predicted that the position Pw is likely to move.
As a result, when the other vehicle 102 turns to the confluence road 106 where the travelable range of the entrance is narrow, it is possible to accurately estimate the possibility of steering in the direction opposite to the direction of the turn to the confluence road 106.

(6) The possibility estimation unit 59 predicts that the closer the road width direction position Pw of the other vehicle 102 is to the roadside 104, the higher the possibility that the road width direction position Pw of the other vehicle 102 will move in the direction away from the roadside 104. To do.
As a result, it is possible to accurately estimate the possibility that the other vehicle 102 will steer in the direction opposite to the direction of turning to the confluence road 106 in an attempt to avoid derailing.

(7) When the width of the traveling lane of the other vehicle 102 is equal to or less than a predetermined value, the possibility estimation unit 59 determines that the position Pw of the other vehicle 102 in the road width direction is likely to move in the direction away from the roadside 104. Predict.
As a result, it is possible to accurately estimate the possibility that the other vehicle 102 will steer in the direction opposite to the direction of turning to the confluence road 106 in an attempt to avoid derailing.

(8) When the motorcycle is present behind the other vehicle 102, the possibility estimation unit 59 predicts that the road width direction position Pw of the other vehicle 102 is likely to move in the direction away from the roadside 104.
As a result, since the other vehicle 102 is traveling closer to the roadside 104 side while being wary of the involvement of the motorcycle, the direction of turning to the confluence road 106 is to avoid derailing when turning to the confluence road 106. The possibility of steering in the opposite direction can be estimated accurately.

(9) Of the angles θ1 and θ2 formed by the traveling road 100 and the confluence road 106, when the angle θ1 close to the other vehicle 102 is an acute angle, the possibility estimation unit 59 moves the other vehicle away from the roadside 104. It is predicted that the position Pw in the road width direction of 102 is likely to move.
As a result, it is possible to accurately estimate the possibility that the other vehicle 102 will steer in the direction opposite to the direction of turning to the confluence road 106 in an attempt to avoid derailing.

(10) When the vehicle length of the other vehicle 102 is determined to be longer than the predetermined length, the possibility estimation unit 59 determines that there is a high possibility that the road width direction position Pw of the other vehicle 102 will move in the direction away from the roadside 104. Predict.
As a result, it is possible to accurately estimate the possibility that the other vehicle 102 will steer in the direction opposite to the direction of turning to the confluence road 106 in an attempt to avoid derailing due to the difference in the inner wheels.

(11) When it is determined that the length of the oncoming vehicle of the other vehicle 102 is longer than the predetermined length, the possibility estimation unit 59 can move the position Pw of the other vehicle 102 in the road width direction in the direction away from the roadside 104. Predict that the sex is low.
As a result, it can be accurately estimated that the possibility that the other vehicle 102 will steer in the direction opposite to the direction in which the other vehicle 102 turns to the confluence road 106 is reduced in order to avoid a collision with an oncoming vehicle having a large vehicle length.

(12) When it is predicted that the road width direction position Pw of the other vehicle 102 may move in the direction away from the roadside 104, the own vehicle route generation unit 37 and the own vehicle 1 and the other vehicle 102 in the road width direction. Generates a target traveling track of the own vehicle 1 that increases the distance of the own vehicle 1. The vehicle control unit 39 makes the own vehicle 1 travel along the target traveling track.
As a result, it is possible to prevent the own vehicle 1 and the other vehicle 102 from being excessively close to each other and realize smooth running.

(13) When it is predicted that the road width direction position Pw of the other vehicle 102 may move in the direction away from the roadside 104, the own vehicle route generation unit 37 generates a speed profile for decelerating the own vehicle 1. .. The vehicle control unit 39 controls the speed of the own vehicle 1 according to the speed profile.
As a result, it is possible to prevent the own vehicle 1 and the other vehicle 102 from being excessively close to each other and realize smooth running.

1 ... own vehicle, 10 ... driving support device, 11 ... object sensor, 12 ... vehicle sensor, 13 ... positioning device, 14 ... map database, 15 ... communication device, 16 ... controller, 17 ... actuator, 18 ... notification device, 21 ... Processor, 22 ... Storage device, 30 ... Object detection unit, 31 ... Own vehicle position estimation unit, 32 ... Map acquisition unit, 33 ... Detection integration unit, 34 ... Object tracking unit, 35 ... Position calculation unit in map, 36 ... Motion prediction unit, 37 ... own vehicle route generation unit, 38 ... vehicle control unit, 39 ... vehicle control unit, 40 ... lane determination unit, 41 ... right / left turn intention prediction unit, 42 ... merging road detection unit, 43 ... merging road width Calculation unit, 44 ... Mobility calculation unit, 45 ... Track prediction unit, 50 ... Confluence road width determination unit, 51 ... Roadside condition determination unit, 52 ... Entrance condition determination unit, 53 ... Road width direction position determination unit, 54 ... Traveling lane width determination unit, 55 ... Two-wheeled vehicle determination unit, 56 ... Angle determination unit, 57 ... Vehicle length determination unit, 58 ... Oncoming vehicle determination unit, 59 ... Possibility estimation unit, 100 ... Travel path, 101 ... Travel path, 102 ... other vehicles, 104 ... roadside, 106 ... confluence road

Claims (14)

Detects the position and direction of travel of other vehicles around the own vehicle,
It is determined whether or not there is a merging road having a predetermined width or less that merges with the traveling path of the other vehicle at a point within a predetermined distance ahead of the traveling direction of the other vehicle.
When it is determined that the merging road exists, it is predicted that the position of the other vehicle in the road width direction may move in a direction away from one of the roadsides of the traveling road facing the merging road. To do,
A method of predicting the behavior of other vehicles. When it is determined that the merging road exists and the direction indicator of the other vehicle is lit, there is a possibility that the position of the other vehicle in the road width direction moves away from one of the roadsides. The behavior prediction method according to claim 1, wherein the presence is predicted. When it is determined that the merging road exists and the other vehicle is decelerating, it is predicted that the position of the other vehicle in the road width direction may move in a direction away from one of the roadsides. The behavior prediction method according to claim 1 or 2, wherein the behavior prediction method is performed. When it is determined that an object having a height equal to or higher than a predetermined height with respect to the traveling road surface or a depression having a depth equal to or higher than a predetermined depth with respect to the traveling road surface exists at the one road end, the one road The behavior prediction method according to any one of claims 1 to 3, wherein it is predicted that the position of the other vehicle in the road width direction is likely to move in a direction away from the end. If there is an obstacle at the entrance of the confluence road that narrows the travelable range in the width direction of the confluence road, the position of the other vehicle in the road width direction may move in a direction away from one of the roadsides. The behavior prediction method according to any one of claims 1 to 4, wherein the behavior is predicted to be high. It is characterized in that it is predicted that the closer the position in the road width direction of the other vehicle is to the one road end, the higher the possibility that the road width direction position of the other vehicle will move in the direction away from the one road end. The behavior prediction method according to any one of claims 1 to 5. When the width of the traveling lane of the other vehicle is equal to or less than a predetermined value, it is predicted that the position of the other vehicle in the road width direction is likely to move in a direction away from one of the roadsides. The behavior prediction method according to any one of Items 1 to 6. Claims 1 to 1, wherein when a motorcycle is present behind the other vehicle, it is predicted that the position of the other vehicle in the road width direction is likely to move in a direction away from one of the roadsides. The behavior prediction method according to any one of 7. When one of the angles formed by the traveling road and the confluence road is an acute angle, the position of the other vehicle in the road width direction can move in a direction away from the one road end. The behavior prediction method according to any one of claims 1 to 8, wherein the property is predicted to be high. A claim characterized in that when it is determined that the length of the other vehicle is longer than a predetermined length, it is predicted that the position of the other vehicle in the road width direction is likely to move in a direction away from one of the roadsides. The behavior prediction method according to any one of Items 1 to 9. When it is determined that the length of the oncoming vehicle of the other vehicle is longer than the predetermined length, it is predicted that it is unlikely that the position of the other vehicle in the road width direction will move in the direction away from the one roadside. The behavior prediction method according to any one of claims 1 to 10. When it is predicted that the position of the other vehicle in the road width direction may move in a direction away from one of the roadsides, the distance between the own vehicle and the other vehicle in the road width direction is increased. Generate a target driving track,
The own vehicle is driven along the target traveling track.
The behavior prediction method according to any one of claims 1 to 11, characterized in that. When it is predicted that the position of the other vehicle in the road width direction may move in the direction away from one of the roadsides, a speed profile for decelerating the own vehicle is generated.
Control the speed of the own vehicle according to the speed profile,
The behavior prediction method according to any one of claims 1 to 12, characterized in that. Sensors that detect objects around your vehicle and
Based on the detection result of the sensor, the position and the traveling direction of another vehicle around the own vehicle are detected, and the vehicle joins the traveling path of the other vehicle at a predetermined distance ahead of the traveling direction of the other vehicle. It is determined whether or not there is a confluence road having a width or less, and when it is determined that the confluence road exists, the roadside of the traveling road is moved away from one of the roadsides facing the confluence road. A controller that predicts that the position of another vehicle in the road width direction may move,
A behavior prediction device for other vehicles, which comprises the above.

Images