JP7149171B2 - Object recognition method and object recognition device - Google Patents

Description

The present invention relates to an object recognition method and an object recognition device.

Conventionally, an object tracking method using a laser radar is known (Patent Document 1). The invention described in Patent Document 1 creates a cluster using detection points detected by a laser radar, calculates the matching degree of the shape of the cluster based on the amount of movement of the detection points included in the cluster, and calculates the degree of matching. A cluster is specified as one object when the matching degree of the shape obtained is equal to or higher than a predetermined threshold. As a result, the invention described in Patent Document 1 can identify an object and also calculate the velocity of the object.

JP 2013-228259 A

In scenes where the own vehicle experiences a large yaw rate (sharp curves, right turns (left turns) at intersections, etc.), the appearance of objects seen from the own vehicle changes significantly. For example, when there is a stopped vehicle ahead of a sharp curve or a parked vehicle ahead of a right turn (left turn) at an intersection, the appearance of the stopped or parked vehicle seen from the own vehicle changes greatly. However, the invention described in Patent Document 1 does not refer to a scene in which a large yaw rate occurs in the own vehicle, and there is a possibility that the velocity of the object cannot be calculated accurately in such a scene.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problem, and an object thereof is to provide an object recognition method and an object recognition apparatus capable of accurately calculating the velocity of an object.

An object recognition method according to an aspect of the present invention estimates a future trajectory of an own vehicle. The object recognition method detects a plurality of first detection points at a first time and detects a plurality of second detection points at a second time different from the first time. The object recognition method uses the distances from the plurality of first detection points to the track, and selects the first detection point closest to the own vehicle, which is the detection point with the shortest distance to the track , from among the plurality of first detection points. , and using the distances from a plurality of second detection points to the track, a detection point closest to the own vehicle, which is the detection point with the shortest distance to the track, is selected from the plurality of second detection points. A second detection point is extracted. The object recognition method calculates the velocity vector of the closest detection point to the own vehicle from the temporal change in the positions of the extracted first detection point and second detection point. The object recognition method decomposes the velocity vector into a tangential velocity component and a normal velocity component of the track, and calculates the tangential velocity component as the relative velocity of the object with respect to the own vehicle.

According to the present invention, it is possible to accurately calculate the velocity of an object.

FIG. 1 is a schematic configuration diagram of an object recognition device according to the first embodiment of the present invention. FIG. 2 is a flow chart explaining an operation example of the object recognition device according to the first embodiment of the present invention. FIG. 3 is a diagram illustrating an example of a driving scene. FIG. 4 is a diagram illustrating an example of a method of classifying detection points. FIG. 5 is a diagram illustrating an example of a method of tracking detection points. FIG. 6 is a diagram illustrating an example of velocity vectors. FIG. 7 is a diagram illustrating an example of decomposition of velocity vectors. FIG. 8 is a schematic configuration diagram of an object recognition device according to a second embodiment of the present invention. FIG. 9 is a flow chart illustrating an operation example of the object recognition device according to the second embodiment of the present invention.

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.

(First embodiment)
(Configuration example of object recognition device 1)
A configuration example of the object recognition device 1 will be described with reference to FIG. As shown in FIG. 1, the object recognition device 1 includes a sensor 10, a map database 11, a GPS receiver 12, a controller 20, an actuator 30, a speaker 31, and a display 32.

The object recognition device 1 may be installed in a vehicle having an automatic driving function, or may be installed in a vehicle without an automatic driving function. Further, the object recognition device 1 may be installed in a vehicle capable of switching between automatic driving and manual driving. Note that automatic driving in this embodiment refers to, for example, a state in which at least one of actuators such as a brake, an accelerator, and a steering is controlled without being operated by a passenger. 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 sensor 10 is a device that is mounted on the own vehicle and detects objects around the own vehicle. The sensor 10 includes lidar, radar, millimeter wave radar, laser range finder, sonar, and the like. The sensor 10 detects moving objects including other vehicles, motorcycles, bicycles and pedestrians, and stationary objects including obstacles, falling objects and parked vehicles as objects around the own vehicle. The sensor 10 also detects the position, orientation (yaw angle), size, speed, acceleration, deceleration, and yaw rate of the moving and stationary objects with respect to the host vehicle. Also, the sensor 10 may include a wheel speed sensor, a steering angle sensor, a gyro sensor, and the like. The sensor 10 outputs detected information to the controller 20 .

In this embodiment, sensor 10 is described as lidar or radar. A lidar or radar scans an object with radio waves and measures the reflected wave to measure the distance and direction to the object. Lidar or radar also acquires point clouds of objects as detection points. A point cloud is a collection of points processed by the controller 20 . A point cloud is usually represented by three-dimensional coordinates (x, y, z). Note that the method of acquiring the point cloud of an object is not limited to lidar or radar. For example, there is a method of obtaining a point cloud of pixels each having a feature distinguishable from surrounding pixels from a camera image. This method includes, for example, the non-patent document "Jianbo Shi and Carlo Tomasi, "Good Features to Track," 1994 IEEE Conference on Computer Vision and Pattern Recognition (CVPR'94), 1994, pp.593-600." method is used.

The map database 11 is a database stored in a car navigation device or the like, and stores various data necessary for route guidance such as road information and facility information. The map database 11 also stores information such as the structure of roads, the number of lanes on roads, the positions of lanes, road boundaries, and targets. The map database 11 outputs map information to the controller 20 in response to requests from the controller 20 . Various data such as road information and target information are not necessarily acquired from the map database 11, and may be acquired by the sensor 10, or may be acquired using vehicle-to-vehicle communication or road-to-vehicle communication. good too. In addition, when various data such as road information and target information are stored in a server installed outside, the controller 20 may acquire these data from the server at any time through communication. Further, the controller 20 may periodically acquire the latest map information from an externally installed server and update the map information it holds.

The GPS receiver 12 detects the position of the vehicle on the ground (hereinafter sometimes referred to as the self-position) and the attitude of the vehicle by receiving radio waves from satellites. The GPS receiver 12 outputs the detected position information and attitude information of the own vehicle to the controller 20 .

The controller 20 is a general-purpose microcomputer having a CPU (Central Processing Unit), memory, and an input/output unit. A computer program for functioning as the object recognition device 1 is installed in the microcomputer. By executing the computer program, the microcomputer functions as a plurality of information processing circuits included in the object recognition device 1. FIG. Here, an example of realizing a plurality of information processing circuits provided in the object recognition apparatus 1 by software will be shown. It is also possible to construct a circuit. Also, a plurality of information processing circuits may be configured by individual hardware. The controller 20 includes a trajectory estimation unit 21, an extraction unit 22, a tracking unit 23, a relative speed calculation unit 24, and a driving support unit 25 as a plurality of information processing circuits.

The trajectory estimation unit 21 estimates the trajectory that the vehicle will travel in the future using the self-position and the attitude of the vehicle.

The extraction unit 22 extracts a detection point closest to the own vehicle from among the detection points detected by the sensor 10 .

The tracking unit 23 tracks the closest detection point to the own vehicle extracted by the extraction unit 22 . Also, the tracking unit 23 calculates the velocity vector of the detection point closest to the own vehicle based on the tracking result.

The relative velocity calculation unit 24 uses the velocity vector calculated by the tracking unit 23 to calculate the relative velocity of the object with respect to the own vehicle.

The driving assistance unit 25 performs driving assistance based on the relative speed calculated by the relative speed calculation unit 24 and the distance L to the detection point closest to the own vehicle. The driving support unit 25 notifies the passenger via the speaker 31 or the display 32 and controls the actuator 30 . The actuator 30 includes a brake actuator, an accelerator actuator, a steering actuator, and the like.

Next, an operation example of the object recognition device 1 will be described with reference to FIGS. 2 to 7. FIG.

In step S101 shown in FIG. 2, the sensor 10 detects an object (stopped vehicle 41 shown in FIG. 3) around the own vehicle 40. FIG. A stopped vehicle 41 shown in FIG. 3 is positioned in front of the own vehicle 40 and is stopped. Sensor 10 detects a plurality of detection points on the surface of stopped vehicle 41 . In addition, in the example shown in FIG. 3, one detection point 60 is shown for convenience of explanation, but a plurality of detection points are actually detected on the surface of the stopped vehicle 41 . A plurality of detection points will be described later. A plurality of detection points may mean a plurality of detection points on the surface of one object, or may mean detection points of various objects around the vehicle 40 .

The process proceeds to step 103 and the GPS receiver 12 detects its own position and the attitude of the vehicle 40 .

The process proceeds to step 105 , and the controller 20 uses the self-position, the attitude of the vehicle 40 and the map database 11 to acquire the lane in which the vehicle 40 is traveling. In addition, below, the lane in which the own vehicle 40 is driving is simply called the own lane. The object recognition apparatus 1 according to the present embodiment will be described as being applied to a scene in which a large yaw rate occurs in the own vehicle 40, such as the curve shown in FIG. Scenes in which a large yaw rate occurs in the own vehicle 40 include curves, right turns (left turns) at intersections, and the like. Of course, the object recognition device 1 according to the present embodiment may be applied to scenes other than scenes in which the host vehicle 40 experiences a large yaw rate.

The process proceeds to step 107, and the trajectory estimation unit 21 estimates the future trajectory 50 of the vehicle 40 on the own lane using the self-position and the attitude of the vehicle 40 (see FIG. 3). As an example, the track 50 passes through the center of the own lane, as shown in FIG.

The process proceeds to step S<b>109 , and the extraction unit 22 extracts the closest detection point to the own vehicle 40 . A detection point closest to the own vehicle 40 will be described with reference to FIG. As shown in FIG. 4, the sensor 10 detects multiple detection points 70 to 73 around the host vehicle 40 . These detection points include detection points related to travel of the own vehicle 40 and detection points not related to the travel of the own vehicle 40 . As an example, a detection point detected on the own lane is determined as a detection point related to travel of the own vehicle 40, and a detection point detected at a position away from the own lane is not related to the travel of the own vehicle 40. It is judged as a detection point.

Among the detection points 70 to 73, the detection points that are not related to the running of the own vehicle 40 are required to be removed as noise. As an example of the method, a method using the distance from the detection point to the trajectory 50 can be considered. The reason is that the shorter the distance from the detection point to the track 50, the higher the possibility that the detection point is related to the running of the own vehicle 40. FIG. In other words, the longer the distance from the detection point to the track 50 , the less likely the detection point is related to the running of the own vehicle 40 .

Therefore, the extraction unit 22 draws perpendicular lines toward the trajectory 50 from the detection points 70 to 73 as shown in FIG. The extraction unit 22 compares the lengths 80 to 83 of the perpendiculars with a predetermined value. In this embodiment, the length 80 is longer than the predetermined value and the lengths 81-83 are less than or equal to the predetermined value. In this case, the extraction unit 22 selects the shortest distance from the lengths 81-83. That is, the extraction unit 22 extracts the detection point with the shortest distance to the trajectory 50 from the detection points 71 to 73 . The detection point with the shortest distance to the track 50 is the closest detection point to the own vehicle 40 . In the example shown in FIG. 4 , the detection point 71 is the detection point closest to the own vehicle 40 . The reason for the comparison with the predetermined value is that detection points 70 positioned longer than the predetermined value are detection points that are not related to the running of the vehicle 40, and such detection points 70 are removed.

The process proceeds to step S111, and the tracking unit 23 tracks the closest detection point to the host vehicle 40 extracted in step S109. Tracking of the detection point closest to own vehicle 40 will be described with reference to FIG. As shown in FIG. 5, at time t, sensor 10 detects a plurality of detection points on the surface of stopped vehicle 41 . In the example shown in FIG. 5, one detection point 60 is shown for convenience of explanation, but actually a plurality of detection points are detected on the surface of the stopped vehicle 41 . The extraction unit 22 extracts the detection point 60 with the shortest distance to the track 50 as the detection point closest to the own vehicle 40 . That is, the detection point 60 is the closest detection point to the own vehicle 40 at the time t.

Time advances, and at time t+1, as shown in FIG. 5, the own vehicle 40 advances a little and the stopped vehicle 41 remains stopped. At time t+1, sensor 10 detects a plurality of detection points (detection points 60 , 61 , 62 ) on the surface of stopped vehicle 41 . The extraction unit 22 extracts the detection point 62 with the shortest distance to the track 50 from among the detection points 60 , 61 , 62 as the detection point closest to the own vehicle 40 .

The detection point closest to the vehicle 40 at time t is the detection point 60 , and the detection point closest to the vehicle 40 at time t+1 is the detection point 62 . When the time advances from time t to time t+1, the tracking unit 23 tracks the detection point 60 and then the detection point 62 in that order. As shown in FIG. 6, the tracking unit 23 tracks the detection point 60 and the detection point 62 in that order, and as a result, a velocity vector 90 directed from the detection point 60 to the detection point 62 is obtained. That is, the tracking unit 23 calculates the velocity vector of the detection point closest to the own vehicle 40 from the temporal change in the positions of the detection points 60 and 62 (transition from the detection point 60 to the detection point 62).

The process proceeds to step S113, and the relative velocity calculator 24 converts the velocity vector 90 calculated in step S111 into velocity component Vt in the tangential direction and velocity component Vn in the normal direction of the track 50, as shown in FIG. disassemble. Then, the relative speed calculator 24 calculates the speed component Vt as the relative speed of the stopped vehicle 41 with respect to the own vehicle 40 .

The process proceeds to step S115, and the driving support unit 25 performs driving support based on the relative speed calculated in step S113 and the distance L (see FIG. 4) to the detection point closest to the own vehicle 40. FIG. For example, when the TTC (Time To Collision) calculated from the relative speed and the distance L is smaller than a predetermined value, the driving support unit 25 notifies the passenger via the speaker 31 or the display 32 or controls the actuator 30. Decelerate.

The process proceeds to step S117, and when the occupant of the own vehicle 40 turns off the ignition (Yes in step S117), the series of processes ends. If the user has not turned off the ignition (No in step S117), the process returns to step S101. Turning off the ignition in this embodiment includes stopping the vehicle 40 and stopping the power supply system of the vehicle 40 . Turning off the ignition may be realized by turning off an ignition switch provided in the vehicle interior of the vehicle 40, or may be realized by turning off a power system switch provided in the vehicle interior of the vehicle 40. good.

(Effect)
As described above, according to the object recognition device 1 according to the first embodiment, the following effects are obtained.

The object recognition device 1 uses the self-position and the attitude of the vehicle 40 to estimate the trajectory 50 along which the vehicle 40 will travel in the future (see FIG. 3). The object recognition device 1 detects a plurality of first detection points (detection points 60 shown in FIG. 5) using the sensor 10 at a first time (time t shown in FIG. 5). The object recognition device 1 detects a plurality of second detection points (detection points 60 to 62 shown in FIG. 5) using the sensor 10 at a second time (time t+1 shown in FIG. 5) different from the first time. To detect. The object recognition device 1 uses the positions of the plurality of first detection points and the trajectory 50 to select the first detection point closest to the own vehicle 40 (the detection point shown in FIG. 5) from the plurality of first detection points. Extract point 60). The object recognition apparatus 1 uses the positions of the plurality of second detection points and the trajectory 50 to select the second detection point closest to the own vehicle 40 (the detection point shown in FIG. 5) from the plurality of second detection points. Extract point 62). The object recognition device 1 calculates the velocity vector of the detection point closest to the own vehicle 40 based on the temporal change in the positions of the extracted first detection point closest to the own vehicle 40 and the second detection point closest to the own vehicle 40. 90 is calculated (see FIG. 6). The object recognition device 1 decomposes the velocity vector 90 into a tangential velocity component Vt and a normal velocity component Vn of the trajectory 50 . The object recognition device 1 calculates the tangential velocity component Vt as the relative velocity of the stopped vehicle 41 with respect to the own vehicle 40 . As a result, the object recognition apparatus 1 can accurately calculate the speed of the stopped vehicle 41 in a scene in which the appearance of the stopped vehicle 41 as seen from the own vehicle 40 changes significantly (a scene in which a large yaw rate occurs in the own vehicle). can. Of course, the object detected by the object recognition device 1 is not limited to the stopped vehicle 41 . The object recognition device 1 can also accurately calculate the speed of a vehicle that is not stopped by the same method.

Further, the object recognition apparatus 1 draws perpendicular lines from the plurality of detection points 70 to 73 toward the track 50, and detects the closest detection points to the own vehicle 40 from among the detection points whose perpendicular lengths 80 to 83 are equal to or less than a predetermined value. Extract points (see FIG. 4). In FIG. 4, lengths 81-83 are equal to or less than a predetermined value. The object recognition device 1 extracts the detection point closest to the own vehicle 40 from among the detection points 71-73 corresponding to the lengths 81-83. Since only detection points within a predetermined range from the track 50 are used in this way, only objects related to the running of the own vehicle 40 are extracted.

Further, the object recognition device 1 extracts the detection point with the shortest perpendicular length as the detection point closest to the own vehicle 40 . In FIG. 4 , the detection point 71 with the shortest perpendicular length 81 is the closest detection point to the own vehicle 40 . As a result, the object recognition device 1 can accurately track the object even if the shape of the road is complicated.

Also, the object recognition device 1 calculates a trajectory 50 based on the map database 11 and the self-position. Thereby, the object recognition device 1 can set the highly accurate trajectory 50 to a long distance. Thereby, the object recognition device 1 can recognize the object earlier.

(Second embodiment)
(Configuration example of object recognition device 2)
Next, a second embodiment of the present invention will be described with reference to FIGS. 8 and 9. FIG. Reference numerals are used for configurations that overlap with the first embodiment, and descriptions thereof are omitted. The following description will focus on the differences.

As shown in FIG. 8 , the object recognition device 2 according to the second embodiment further includes a camera 13 and a cluster extraction section 26 .

The camera 13 is mounted on the own vehicle 40 and photographs the surroundings of the own vehicle 40 . The camera 13 has an imaging device such as a CCD (charge-coupled device) or a CMOS (complementary metal oxide semiconductor). The camera 13 has an image processing function, and detects white lines, road boundaries, road edges, targets (for example, curbs), etc. from captured images (camera images).

The cluster extraction unit 26 extracts a plurality of detection points as clusters (objects). Such a method is called clustering. As an example, the cluster extraction unit 26 clusters detection points whose distance between two points is equal to or less than a predetermined value from among the plurality of detection points. In the example shown in FIG. 5, the detection points 60 to 62 are clustered at time t+1 and extracted as a cluster indicating the stopped vehicle 41 .

Next, an operation example of the object recognition device 2 will be described with reference to the flowchart of FIG.

Steps S201 to 203, 207 and 213 to 219 are the same as steps S101 to 103, 107 and 111 to 117 shown in FIG. 2, so description thereof will be omitted.

In step S205, the controller 20 acquires the lane (own lane) in which the own vehicle 40 is traveling using the self position and the camera image. In the second embodiment, the controller 20 acquires the own lane using camera images instead of the map database 11, so even in areas where the road structure is not reflected in the map database 11, the own lane can be acquired. .

In step S209, the cluster extracting unit 26 extracts detection points whose distance to the trajectory 50 is equal to or less than a predetermined value, as in the first embodiment, from the plurality of detection points detected in step S201. From the detection points thus extracted, the cluster extraction unit 26 clusters the detection points whose distance between two points is equal to or less than a predetermined value. Thereby, the cluster extraction unit 26 can extract clusters close to the trajectory 50 .

In step S<b>211 , the extraction unit 22 extracts the detection point with the shortest distance to the track 50 as the detection point closest to the own vehicle 40 from among the detection points included in the clusters extracted in step S<b>209 .

(Effect)
As described above, according to the object recognition device 2 according to the second embodiment, the following effects are obtained.

As in the first embodiment, the object recognition device 2 draws perpendicular lines from a plurality of detection points toward the trajectory 50, and extracts detection points whose perpendicular lengths are equal to or less than a predetermined value. From the detection points thus extracted, the object recognition device 2 clusters the detection points whose distance between the two points is equal to or less than a predetermined value. Thereby, the object recognition device 2 can extract clusters close to the trajectory 50 . Then, the object recognition device 2 extracts the detection point with the shortest distance to the trajectory 50 from among the detection points included in the clustered clusters as the detection point closest to the own vehicle 40 . In this way, after clustering the detection points, the object recognition device 2 extracts the detection points closest to the own vehicle 40, so noise can be removed and objects can be recognized with high accuracy.

The object recognition device 2 calculates the trajectory 50 based on the road structure in front of the own vehicle 40 acquired by the camera 13 . As a result, the object recognition device 2 can acquire its own lane even in an area where the road structure is not reflected in the map database 11 .

Each function described in the above embodiments may be implemented by one or more processing circuits. Processing circuitry includes programmed processing devices, such as processing devices that include electrical circuitry. Processing circuitry also includes devices such as application specific integrated circuits (ASICs) and circuit components arranged to perform the described functions. Also, the object recognition device 1 and the object recognition device 2 can improve the function of the computer.

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, 2 Object recognition device 10 Sensor 11 Map database 12 GPS receiver 13 Camera 20 Controller 21 Trajectory estimation unit 22 Extraction unit 23 Tracking unit 24 Relative velocity calculation unit 25 Driving support unit 26 Cluster extraction unit 30 Actuator 31 Speaker 32 Display

Claims (7)

A sensor mounted on the own vehicle is used to detect a plurality of detection points on the surface of an object around the own vehicle, and the object is recognized using the plurality of detection points detected by the sensor. An object recognition method comprising:
estimating the future trajectory of the own vehicle;
Detecting the plurality of first detection points using the sensor at a first time;
Detecting a plurality of second detection points using the sensor at a second time different from the first time,
Using the distances from the plurality of first detection points to the trajectory, a first detection point closest to the subject vehicle, which is the detection point with the shortest distance to the trajectory, is selected from among the plurality of first detection points. Extract the detection points of
Using the distances from the plurality of second detection points to the trajectory, a second detection point closest to the own vehicle, which is the detection point with the shortest distance to the trajectory, is selected from the plurality of second detection points. Extract the detection points of
calculating the velocity vector of the detection point closest to the own vehicle from the time change of the extracted positions of the first detection point closest to the own vehicle and the second detection point closest to the own vehicle;
decomposing the velocity vector into a tangential velocity component and a normal velocity component of the trajectory;
An object recognition method, wherein the speed component in the tangential direction is calculated as a relative speed of the object with respect to the own vehicle. drawing perpendicular lines from the plurality of detection points toward the trajectory;
2. The object recognition method according to claim 1, wherein a detection point closest to the own vehicle is extracted from among the detection points whose vertical length is equal to or less than a predetermined value. 3. The object recognition method according to claim 2, wherein the detection point having the shortest perpendicular length is extracted as the detection point closest to the host vehicle. drawing perpendicular lines from the plurality of detection points toward the trajectory;
clustering the detection points whose distance between the two points is a predetermined value or less from among the detection points whose perpendicular length is a predetermined value or less;
4. The object recognition method according to any one of claims 1 to 3, wherein a detection point closest to the own vehicle is extracted from among the detection points included in the clustered clusters. The object recognition method according to any one of claims 1 to 4, wherein the trajectory is calculated based on a map database and the position of the own vehicle. The object recognition according to any one of claims 1 to 4, wherein the trajectory is calculated based on a road structure in front of the vehicle, which is acquired by a camera mounted on the vehicle. Method. A sensor mounted on the own vehicle is used to detect a plurality of detection points on the surface of an object around the own vehicle, and the object is recognized using the plurality of detection points detected by the sensor. An object recognition device comprising a controller,
The controller is
estimating the future trajectory of the own vehicle;
Detecting the plurality of first detection points using the sensor at a first time;
Detecting a plurality of second detection points using the sensor at a second time different from the first time,
Using the distances from the plurality of first detection points to the trajectory, a first detection point closest to the subject vehicle, which is the detection point with the shortest distance to the trajectory, is selected from among the plurality of first detection points. Extract the detection points of
Using the distances from the plurality of second detection points to the trajectory, a second detection point closest to the own vehicle, which is the detection point with the shortest distance to the trajectory, is selected from the plurality of second detection points. Extract the detection points of
calculating the velocity vector of the detection point closest to the own vehicle from the time change of the extracted positions of the first detection point closest to the own vehicle and the second detection point closest to the own vehicle;
decomposing the velocity vector into a tangential velocity component and a normal velocity component of the trajectory;
An object recognition device, wherein the speed component in the tangential direction is calculated as a relative speed of the object with respect to the own vehicle.

Images