Obstacle avoidance path dynamic planning method considering speed nonlinear change

Abstract

The invention discloses an obstacle avoidance path dynamic planning method considering speed nonlinear change, and relates to the technical field of vehicle collision detection and avoidance, and the method comprises the following steps: detecting the current positions and speed vectors of a target vehicle and an obstacle in real time based on the vehicle detection and network connection technology; on the basis of the current position and the velocity vector, predicting the velocity vector of the obstacle at the next moment, and obtaining movement track coordinates of the target vehicle and the obstacle; based on the movement track coordinates, taking the obstacle as a center, and combining the sizes of the target vehicle and the obstacle to obtain a relative conflict area of the target vehicle and the obstacle; based on the relative conflict area, constructing a target vehicle collision speed set; and based on the target vehicle collision speed set, the target vehicle collision risk is judged, and the collision avoidance speed of the target vehicle is obtained. The collision risk of the target vehicle and the obstacle at any moment can be accurately detected in real time, and the problem that an existing method has large errors is solved.

Description

Technical Field

[0001] The invention relates to the technical field of vehicle collision detection and avoidance, and in particular to a dynamic planning method for an obstacle avoidance path taking into account nonlinear changes in speed. Background Art

[0002] Traffic accidents caused by collisions between driving vehicles and other motor vehicles, non-motor vehicles, pedestrians, debris, etc. have always been a key issue of social concern. How to detect and avoid vehicle collisions in advance, actively and effectively has always been a hot research area for many scholars. At present, the vehicle collision detection method focuses on directly measuring the time when the target vehicle and other obstacles reach the collision point between them. The dynamic changes of the obstacle speed and trajectory are not fully considered. The collision point obtained by treating the vehicle as a particle ignores the influence of the vehicle size, resulting in large errors in vehicle collision detection and greatly restricting the accurate selection of the vehicle collision avoidance speed. With the rapid development of vehicle identification, detection and networking technologies, real-time detection of traffic operation status, extraction and transmission of vehicle dynamic trajectory data, and high-precision vehicle driving and control have gradually become possible. Therefore, based on the real-time vehicle detection information, the dynamic changes of the position and speed of obstacles around the target vehicle are predicted, and then the collision area is constructed by considering the size of the target vehicle and the obstacle. Based on the current driving state of the target vehicle, the area range and speed relationship of the possible collision with the obstacle at the future moment are judged, and then the vehicle collision risk is judged, and the vehicle driving state is adjusted to avoid collision, which can significantly improve the rationality and accuracy of vehicle collision detection and avoidance. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a dynamic planning method for obstacle avoidance paths taking into account nonlinear changes in speed, which solves the problem of large errors in the prior art.

[0004] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: a dynamic planning method for obstacle avoidance path considering nonlinear changes in speed, comprising the following steps:

[0005] S1: Based on vehicle detection and networking technology, the current position and velocity vector of the target vehicle and obstacles are detected in real time;

[0006] S2: Based on the current position and velocity vector, predict the velocity vector of the obstacle at the next moment, and obtain the motion trajectory coordinates of the target vehicle and the obstacle;

[0007] S3: Based on the motion trajectory coordinates, with the obstacle as the center, combined with the size of the target vehicle and the obstacle, the relative conflict area between the target vehicle and the obstacle is obtained;

[0008] S4: construct a target vehicle collision speed set based on the relative conflict area;

[0009] S5: Based on the target vehicle collision speed set, the target vehicle collision risk is determined, and the collision avoidance speed of the target vehicle is obtained, and the dynamic planning of the obstacle avoidance path considering the nonlinear change of speed is completed.

[0010] The beneficial effect of the above scheme is: the present invention provides a vehicle collision detection and avoidance method taking into account the dynamic changes in the speed of an obstacle. The method determines the possible area and speed of a target vehicle colliding with the obstacle by taking into account the dynamic changes in the speed of the obstacle, and then judges the risk of collision and determines the collision avoidance method. This can effectively improve the initiative, flexibility, accuracy and efficiency of vehicle collision risk and avoidance, and solves the problem of large errors in existing methods.

[0011] Furthermore, the velocity vector of the obstacle in S2 at the next moment is:

[0012]

[0013] in, and are the velocity vectors of the obstacle at time k+i and time k, respectively, and a O is the obstacle acceleration vector, is the time difference from time k to time k+i.

[0014] The beneficial effect of the above further solution is: through the above technical solution, the velocity vector at time k+i is predicted based on the obstacle position and vector velocity acquired in real time.

[0015] Furthermore, the motion trajectory coordinates of the target vehicle and the obstacle in S2 are:

[0016] The motion trajectory coordinates of the target vehicle at time k are:

[0017] The motion trajectory coordinates of the target vehicle at time k+i are:

[0018] The coordinates of the obstacle's motion trajectory at time k are:

[0019] The coordinates of the obstacle's motion trajectory at time k+i are:

[0020]

[0021]

[0022]

[0023]

[0024] in, and are the x-axis and y-axis direction components of the target vehicle’s velocity at time k+i, and are the x-axis and y-axis components of the obstacle's velocity at time k+i, respectively, and a O,x and a O,y are the x-axis and y-axis direction components of the acceleration of the obstacle at time k+i respectively;

[0025]

[0026]

[0027]

[0028]

[0029] a O,x =a O sinθ

[0030] a O,y =a O cosθ

[0031] Among them, α is the angle between the target vehicle's moving direction and the x-axis, θ is the angle between the obstacle's moving direction and the x-axis, is the velocity vector of the target vehicle at time k+i.

[0032] The beneficial effect of the above further scheme is: through the above technical scheme, the position coordinates of the target vehicle and the obstacle at time k+i are calculated according to the positions of the target vehicle and the obstacle at time k, so as to realize the determination of the motion trajectory coordinates.

[0033] Furthermore, the relative conflict area in S3 is determined by constructing a circular area with the sum of the radius of the target vehicle and the obstacle as the circle radius and the center of the obstacle as the circle point based on the motion trajectory and velocity vector of the target vehicle and the obstacle at time k+i. Then, the relative conflict area RCC between the target vehicle and the obstacle at time k+i is:

[0034] RCC=P O ⊕ConfP(O)={P C |||P C -P O ||≤r C +r O =R}

[0035] Among them, P O and P Care the centroid positions of the obstacle and the target vehicle, respectively. ConfP(O) is the circular area with the centroid of the obstacle as the circle point. ⊕ is the Minkowski vector sum operation. C and r O are the radii of the target vehicle and obstacle respectively, and R is the sum of the radii of the target vehicle and obstacle.

[0036] The beneficial effect of the above further scheme is: through the above technical scheme, based on the constructed velocity vector linear model and motion trajectory model, the position and appropriate speed of the target vehicle and the obstacle at time k+i are obtained, and the relative conflict area is determined by constructing a circular area with the sum of the target vehicle radius and the obstacle radius as the radius and the obstacle centroid as the center.

[0037] Furthermore, constructing the target vehicle collision speed set in S4 includes the following steps:

[0038] S4-1: When the target vehicle collides with the obstacle at time k+i, the position coordinates of the target vehicle fall within the relative conflict area, and the formula is:

[0039] P C (t k+i )∈P O (t k+i )⊕ConfP(O)

[0040] Among them, P C (·) is the target vehicle position coordinate, P O (·) is the obstacle position coordinate, t k+i is the k+i moment;

[0041] S4-2: When the position coordinates of the target vehicle fall within the relative conflict area, the relationship between the target vehicle and the relative conflict area at time k is obtained as follows:

[0042] P C (t k )+v c (t k+i -t k )∈P O (t k+i )⊕ConfP(O)

[0043] Among them, t k is the k moment, v c is the target vehicle velocity vector;

[0044] S4-3: Based on the relationship between the target vehicle at time k and the relative conflict area, the speed set VO at which the target vehicle collides with the obstacle at time k+i is obtained as follows:

[0045]

[0046] The beneficial effect of the above further solution is: through the above technical solution, based on the relative conflict area, according to the relationship between the target vehicle position and the relative conflict area, a collision speed set is constructed.

[0047] Furthermore, the target vehicle collision risk determination method in S5 is: according to the velocity vector of the target vehicle at time k+i The collision speed set VO is used to determine whether the target vehicle will collide with the obstacle, specifically:

[0048] (1) If Then the target vehicle will collide with the obstacle at time k+i;

[0049] (2) If Then the target vehicle will not collide with the obstacle at time k+i.

[0050] The beneficial effect of the above further solution is: by judging the relationship between the target vehicle velocity vector and the vehicle collision velocity set, the collision risk of the target vehicle is judged.

[0051] Further, obtaining the collision avoidance speed of the target vehicle in S5 includes the following steps:

[0052] S5-1: When the target vehicle collides with the obstacle at time k+i, the vector speed of the target vehicle at time k is adjusted to:

[0053]

[0054] in, is the velocity vector of the target vehicle at time k, a C is the maximum vector acceleration that the target vehicle can achieve based on its own performance, is the time difference from time k to time k+i;

[0055] S5-2: Based on the adjustment result of the target vehicle vector speed, by considering the principle of balancing driving safety and efficiency, multiple acceleration vectors of the target vehicle with different values ​​and directions at time k+i are obtained. The formula is:

[0056] F(v)=k1f1+k2f2

[0057]

[0058]

[0059] Among them, F(v) is the evaluation function for adjusting the vector speed, f1 is the sub-function representing efficiency, f2 is the sub-function representing safety, k1 and k2 are constant coefficients, and v Eis the expected speed of the target vehicle pointing to the target location, is the velocity component of the candidate vector velocity in the desired direction, To select the lth alternative speed, the Euclidean distance between the target vehicle and the obstacle at time k+i;

[0060] S5-3: Compare the evaluation function of the adjusted vector speed with the evaluation indexes of the candidate speeds, and select the adjusted vector speed with the best evaluation index as the collision avoidance speed of the target vehicle.

[0061] The beneficial effect of the above further scheme is: based on the expected driving direction of the target vehicle and taking into account the principles of vehicle driving safety and efficiency, the speed of the target vehicle to avoid collision is determined through the above technical scheme. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a flow chart of a dynamic planning method for obstacle avoidance paths that takes into account nonlinear changes in speed.

[0063] Figure 2 A schematic diagram is provided to represent the vector position and vector velocity distribution of conflicting vehicles. DETAILED DESCRIPTION

[0064] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0065] like Figure 1 As shown, a dynamic planning method for obstacle avoidance path considering nonlinear speed change includes the following steps:

[0066] S1: Based on vehicle detection and networking technology, the current position and velocity vector of the target vehicle and obstacles are detected in real time;

[0067] S2: Based on the current position and velocity vector, predict the velocity vector of the obstacle at the next moment, and obtain the motion trajectory coordinates of the target vehicle and the obstacle;

[0068] S3: Based on the motion trajectory coordinates, with the obstacle as the center, combined with the size of the target vehicle and the obstacle, the relative conflict area between the target vehicle and the obstacle is obtained;

[0069] S4: construct a target vehicle collision speed set based on the relative conflict area;

[0070] S5: Based on the target vehicle collision speed set, the target vehicle collision risk is determined, and the collision avoidance speed of the target vehicle is obtained, and the dynamic planning of the obstacle avoidance path considering the nonlinear change of speed is completed.

[0071] The present invention is mainly used for detecting and avoiding collision risks between vehicles and other motor vehicles, non-motor vehicles, pedestrians, sundries, etc. Considering that obstacles around the vehicle (other motor vehicles, non-motor vehicles, pedestrians, sundries, etc.) are mostly traveling at dynamically changing speeds (such as acceleration / deceleration), and vehicle collisions do not occur at a certain point but within a certain possible range (such as the body contact range between the target vehicle and the obstacle), the present invention provides a dynamic planning method for obstacle avoidance paths that takes into account nonlinear changes in speed. Based on the real-time detected position and vector velocity of the target vehicle and obstacle, the law of change of the obstacle vector velocity is determined. Based on the position and vector velocity geometric relationship between the target vehicle and the obstacle, the conflict area and speed where the target vehicle and the obstacle may collide are determined, and then the collision risk between the target vehicle and the obstacle is judged, and the target vehicle vector velocity is actively adjusted to avoid collision. The specific steps are as follows:

[0072] In one embodiment of the present invention, traffic state detection: based on vehicle detection and networking technology, a target vehicle finds a uniformly speeding obstacle around it during driving. The position and vector speed detection results of the target vehicle and the obstacle are shown in Table 1:

[0073] Table 1 Target vehicle and obstacle trajectory detection

[0074]

[0075] Velocity vector prediction: Based on the obstacle position and vector velocity obtained in real time, it is known that the obstacle is moving at a uniform speed, and its acceleration vector a O for:

[0076]

[0077] Therefore, the velocity vector of the obstacle at time k+i can be obtained:

[0078] The velocity vector of the obstacle in S2 at the next moment is:

[0079]

[0080] in, and are the velocity vectors of the obstacle at time k+i and time k, respectively, and a O is the obstacle acceleration vector, is the time difference from time k to time k+i.

[0081] Determination of motion trajectory coordinates: Based on the positions of the target vehicle and the obstacle at time k, according to the following motion equation, the position coordinates of the target vehicle and the obstacle at time k+i are obtained respectively:

[0082] The motion trajectory coordinates of the target vehicle and the obstacle in S2 are:

[0083] The motion trajectory coordinates of the target vehicle at time k are:

[0084] The motion trajectory coordinates of the target vehicle at time k+i are:

[0085] The coordinates of the obstacle's motion trajectory at time k are:

[0086] The coordinates of the obstacle's motion trajectory at time k+i are:

[0087]

[0088]

[0089]

[0090]

[0091] in, and are the x-axis and y-axis direction components of the target vehicle’s velocity at time k+i, and are the x-axis and y-axis components of the obstacle's velocity at time k+i, respectively, and a O,x and a O,y are the x-axis and y-axis direction components of the acceleration of the obstacle at time k+i respectively;

[0092]

[0093]

[0094]

[0095]

[0096]

[0097] a O,y =a O cosθ

[0098] Among them, α is the angle between the target vehicle's moving direction and the x-axis, θ is the angle between the obstacle's moving direction and the x-axis, is the velocity vector of the target vehicle at time k+i.

[0099] like Figure 2As shown in FIG. 1 , the method for determining the relative conflict area in S3 is as follows: based on the motion trajectory and velocity vector of the target vehicle and the obstacle at time k+i, a circular area with the sum of the radius of the target vehicle and the obstacle as the circle radius and the center of the obstacle as the circle point is constructed. Then, the relative conflict area RCC between the target vehicle and the obstacle at time k+i is:

[0100] RCC=P O ⊕ConfP(O)={P C |||P C -P O ||≤r C +r O =R}

[0101] Among them, P O and P C are the centroid positions of the obstacle and the target vehicle, respectively. ConfP(O) is the circular area with the centroid of the obstacle as the circle point. ⊕ is the Minkowski vector sum operation. C and r O are the radii of the target vehicle and obstacle respectively, and R is the sum of the radii of the target vehicle and obstacle.

[0102] In S4, a target vehicle collision speed set is constructed, including the following steps:

[0103] S4-1: When the target vehicle collides with the obstacle at time k+i, the position coordinates of the target vehicle fall within the relative conflict area, and the formula is:

[0104] P C (t k+i )∈P O (t k+i )⊕ConfP(O)

[0105] Among them, P C (·) is the target vehicle position coordinate, P O (·) is the obstacle position coordinate, t k+i is the k+i moment;

[0106] S4-2: When the position coordinates of the target vehicle fall within the relative conflict area, the relationship between the target vehicle and the relative conflict area at time k is obtained as follows:

[0107] P C (t k )+v c (t k+i -t k )∈P O (t k+i )⊕ConfP(O)

[0108] Among them, t kis the k moment, v c is the target vehicle velocity vector;

[0109] S4-3: Based on the relationship between the target vehicle at time k and the relative conflict area, the speed set VO at which the target vehicle collides with the obstacle at time k+i is obtained as follows:

[0110]

[0111] The target vehicle collision risk judgment method in S5 is: according to the velocity vector of the target vehicle at time k+i The collision speed set VO is used to determine whether the target vehicle will collide with the obstacle, specifically:

[0112] (1) If Then the target vehicle will collide with the obstacle at time k+i;

[0113] (2) If Then the target vehicle will not collide with the obstacle at time k+i.

[0114] Vehicle collision risk judgment: Compare the speed v of the target vehicle at time k+i c With the vehicle collision speed set VO, it is found It is known that the target vehicle will collide with the obstacle at time k+i.

[0115] Obtaining the collision avoidance speed of the target vehicle in S5 includes the following steps:

[0116] S5-1: When the target vehicle collides with the obstacle at time k+i, the vector speed of the target vehicle at time k is adjusted to:

[0117]

[0118] in, is the velocity vector of the target vehicle at time k, a C is the maximum vector acceleration that the target vehicle can achieve based on its own performance, is the time difference from time k to time k+i;

[0119] S5-2: Based on the adjustment result of the target vehicle vector speed, by considering the principle of balancing driving safety and efficiency, multiple acceleration vectors of the target vehicle with different values ​​and directions at time k+i are obtained. The formula is:

[0120] F(v)=k1f1+k2f2

[0121]

[0122]

[0123] Among them, F(v) is the evaluation function for adjusting the vector speed, f1 is the sub-function representing efficiency, f2 is the sub-function representing safety, k1 and k2 are constant coefficients, and v E is the expected speed of the target vehicle pointing to the target location, is the velocity component of the candidate vector velocity in the desired direction, To select the lth alternative speed, the Euclidean distance between the target vehicle and the obstacle at time k+i;

[0124] S5-3: Compare the evaluation function of the adjusted vector speed with the evaluation indexes of the candidate speeds, and select the adjusted vector speed with the best evaluation index as the collision avoidance speed of the target vehicle.

[0125] Collision avoidance speed selection: According to the target vehicle's own performance, its maximum vector acceleration is a C , the coefficients k1 and k2 of the safety and efficiency sub-items in the evaluation function of the target vehicle are a and b respectively. Then the target vehicle is at the collision avoidance speed, and the optional vector speeds of the target vehicle at time k+i and the evaluation indexes of each vector speed are shown in Table 2:

[0126] Table 2 Alternative vector speeds and speed evaluation indexes of target vehicles

[0127]

[0128] Comparing the evaluation index of each alternative speed, we know:

[0129] G4>G2>G1>G6>G7>G3>G5

[0130] So choose speed As the target vehicle's speed at time k+i.

[0131] The present invention makes up for the defect of assuming that the speed of obstacles is constant by considering the dynamic changes of the speed of obstacles around the target vehicle; by considering the size of the target vehicle and the obstacle, the area where the two collide is determined, which makes up for the defect of assuming that the collision occurs at a certain conflict point; by calculating the vehicle collision speed from the collision area, a more reasonable alternative speed is provided for the target vehicle to avoid collision. In summary, the present invention can effectively improve the initiative, flexibility, accuracy and efficiency of vehicle collision detection and avoidance speed selection, thereby significantly reducing the risk of vehicle collision and improving road safety.

[0132] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the invention.

Claims

1. A dynamic planning method for obstacle avoidance path considering nonlinear changes in speed, characterized in that: The following steps are involved: S1: Based on vehicle detection and networking technology, the current position and velocity vector of the target vehicle and obstacles are detected in real time; S2: Based on the current position and velocity vector, predict the velocity vector of the obstacle at the next moment, and obtain the motion trajectory coordinates of the target vehicle and the obstacle; S3: Based on the motion trajectory coordinates, with the obstacle as the center, combined with the size of the target vehicle and the obstacle, the relative conflict area between the target vehicle and the obstacle is obtained; S4: construct a target vehicle collision speed set based on the relative conflict area; S5: Based on the target vehicle collision speed set, the target vehicle collision risk is determined, and the collision avoidance speed of the target vehicle is obtained, and the dynamic planning of the obstacle avoidance path considering the nonlinear change of speed is completed.

2. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 1, characterized in that: The velocity vector of the obstacle in S2 at the next moment is: in, and are the velocity vectors of the obstacle at time k+i and time k, respectively, and a O is the obstacle acceleration vector, is the time difference from time k to time k+i.

3. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 2, characterized in that: The motion trajectory coordinates of the target vehicle and the obstacle in S2 are: The motion trajectory coordinates of the target vehicle at time k are: The motion trajectory coordinates of the target vehicle at time k+i are: The coordinates of the obstacle's motion trajectory at time k are: The coordinates of the obstacle's motion trajectory at time k+i are: in, and are the x-axis and y-axis direction components of the target vehicle’s velocity at time k+i, and are the x-axis and y-axis components of the obstacle's velocity at time k+i, respectively, and a O,x and a O,y are the x-axis and y-axis direction components of the acceleration of the obstacle at time k+i respectively; a O,x =a O sinθ a O,y =a O cosθ Among them, α is the angle between the target vehicle's moving direction and the x-axis, θ is the angle between the obstacle's moving direction and the x-axis, is the velocity vector of the target vehicle at time k+i.

4. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 1, characterized in that: The method for determining the relative conflict area in S3 is: based on the motion trajectory and velocity vector of the target vehicle and the obstacle at time k+i, a circular area with the sum of the radius of the target vehicle and the obstacle as the circle radius and the center of the obstacle as the circle point is constructed. Then, the relative conflict area RCC between the target vehicle and the obstacle at time k+i is: Among them, P O and P C are the centroid positions of the obstacle and the target vehicle, respectively. ConfP(O) is the circular area with the centroid of the obstacle as the point. is the Minkowski vector sum operation, r C and r O are the radii of the target vehicle and obstacle respectively, and R is the sum of the radii of the target vehicle and obstacle.

5. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 4, characterized in that: In S4, a target vehicle collision speed set is constructed. The following steps are included: S4-1: When the target vehicle collides with the obstacle at time k+i, the position coordinates of the target vehicle fall within the relative conflict area, and the formula is: Among them, P C (·) is the target vehicle position coordinate, P O (·) is the obstacle position coordinate, t k+i is the k+i moment; S4-2: When the position coordinates of the target vehicle fall within the relative conflict area, the relationship between the target vehicle and the relative conflict area at time k is obtained as follows: Among them, t k is the k moment, v c is the target vehicle velocity vector; S4-3: Based on the relationship between the target vehicle at time k and the relative conflict area, the speed set VO at which the target vehicle collides with the obstacle at time k+i is obtained as follows:

6. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 5, characterized in that: The target vehicle collision risk determination method in S5 is: according to the velocity vector of the target vehicle at time k+i The collision speed set VO is used to determine whether the target vehicle will collide with the obstacle, specifically: (1) If Then the target vehicle will collide with the obstacle at time k+i; (2) If Then the target vehicle will not collide with the obstacle at time k+i.

7. The obstacle avoidance path dynamic planning method considering nonlinear speed change according to claim 6, characterized in that: Obtaining the collision avoidance speed of the target vehicle in S5 includes the following steps: S5-1: When the target vehicle collides with the obstacle at time k+i, the vector speed of the target vehicle at time k is adjusted to: in, is the velocity vector of the target vehicle at time k, a C is the maximum vector acceleration that the target vehicle can achieve based on its own performance, is the time difference from time k to time k+i; S5-2: Based on the adjustment result of the target vehicle vector speed, by considering the principle of balancing driving safety and efficiency, multiple acceleration vectors of the target vehicle with different values ​​and directions at time k+i are obtained. The formula is: F(v)=k1f1+k2f2 Among them, F(v) is the evaluation function for adjusting the vector speed, f1 is the sub-function representing efficiency, f2 is the sub-function representing safety, k1 and k2 are constant coefficients, and v E is the expected speed of the target vehicle pointing to the target location, is the velocity component of the candidate vector velocity in the desired direction, To select the lth alternative speed, the Euclidean distance between the target vehicle and the obstacle at time k+i; S5-3: Compare the evaluation function of the adjusted vector speed with the evaluation indexes of the candidate speeds, and select the adjusted vector speed with the best evaluation index as the collision avoidance speed of the target vehicle.

Images