Implementation method for avoiding adjacent vehicle, vehicle and terminal equipment

Abstract

The invention relates to the technical field of intelligent driving, in particular to an implementation method for avoiding an adjacent vehicle, a vehicle and terminal equipment. The method comprises the following steps: determining whether to avoid the obstacle of the adjacent vehicle according to the transverse relative distance between the vehicle and the adjacent vehicle and a calibrated safety distance threshold value, or according to the transverse relative speed between the vehicle and the adjacent vehicle and a calibrated collision time threshold value; if an obstacle avoidance adjacent vehicle is determined, calculating all control points corresponding to the multiple sections of third-order Bezier curves according to the transverse speed at a planned expected end point and the speed increment of each planned road section, and according to the transverse offset between the initial position of obstacle avoidance and the expected end point and the distance increment of each planned road section; obtaining a multi-section third-order Bezier curve based on all the control points; each section of the third-order Bezier curve corresponds to one planned road section; and respectively calculating a corresponding planned road section based on each section of the third-order Bezier curve, and further controlling the self-vehicle to advance according to each planned road section so as to avoid an adjacent vehicle which collides with an adjacent lane of the self-vehicle and runs in the same direction.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular, to a method for implementing avoiding adjacent vehicles, a vehicle, and a terminal device. Background Art

[0002] With the continuous increase in traffic flow, the incidence of traffic accidents is also rising. Although traditional passive safety measures (such as airbags, anti-lock braking systems) can reduce injuries after an accident, they cannot prevent accidents from occurring. Therefore, currently, vehicles are provided with systems such as AEB (Automatic Emergency Braking), LDP (Lane Departure Prevention), and ELK (Emergency Lane Keeping Assist) to actively avoid obstacles.

[0003] However, currently, the AEB system mainly brakes longitudinally to avoid obstacles in front; the LDP system mainly corrects the deviation of unconscious lane crossing; the ELK system mainly corrects the deviation actively for oncoming vehicles in the adjacent lane, the road edge, and vehicles coming from behind.

[0004] However, accidents where a vehicle in the same direction in the adjacent lane hits the vehicle itself occur frequently, and the above AEB, LDP, and ELK systems cannot prevent such accidents from occurring. Summary of the Invention

[0005] In view of this, embodiments of the present application provide a method for implementing avoiding adjacent vehicles, a vehicle, and a terminal device, which can effectively solve the problem that the vehicle cannot avoid in time when a vehicle in the same direction in the adjacent lane hits the vehicle itself in the prior art, etc.

[0006] In a first aspect, an embodiment of the present application provides a method for implementing avoiding adjacent vehicles, including:

[0007] Determine whether to avoid the adjacent vehicle according to the lateral relative distance between the vehicle itself and the adjacent vehicle and the calibrated safety distance threshold, or according to the lateral relative speed between the vehicle itself and the adjacent vehicle and the calibrated collision time threshold;

[0008] If it is determined to avoid the adjacent vehicle, calculate all control points corresponding to multiple cubic Bézier curves according to the lateral speed at the planned expected end point and the speed increment of each planned section, and according to the lateral offset between the initial position of the avoidance and the expected end point and the distance increment of each planned section, and then obtain multiple cubic Bézier curves based on all the control points; wherein, each cubic Bézier curve corresponds to a planned section;

[0009] Calculate the corresponding planned road segments based on each cubic Bézier curve, and then control the ego vehicle to move forward according to each planned road segment to avoid adjacent vehicles.

[0010] In some embodiments, all control points corresponding to multiple cubic Bézier curves are calculated according to the lateral speed at the planned desired end point, the speed increment of each planned road segment, and according to the lateral offset between the initial position of obstacle avoidance and the desired end point and the distance increment of each planned road segment, including:

[0011] All control points corresponding to multiple cubic Bézier curves are calculated according to the lateral speed at the planned desired end point, according to the planned time and acceleration of each planned road segment, and according to the lateral offset between the initial position of obstacle avoidance and the desired end point and the planned time, initial lateral speed and increment coefficient of each planned road segment; wherein, the acceleration is determined according to the control points of the corresponding cubic Bézier curve; the increment coefficient is determined according to the control points of the corresponding cubic Bézier curve.

[0012] In some embodiments, the planned times of the multiple cubic Bézier curves are set to T1, T2, …, T i ; i is the total number of planned road segments;

[0013] All control points corresponding to multiple cubic Bézier curves are calculated according to the lateral speed at the planned desired end point, according to the planned time and acceleration of each planned road segment, and according to the lateral offset between the initial position of obstacle avoidance and the desired end point and the planned time, initial lateral speed and increment coefficient of each planned road segment, including calculating all control points using the following formula:

[0014] T1a1 + T2a2 + … + T i a i = 0

[0015] (T1 2 L1 + T1v0) + [(T2) 2 L2 + T2(T1a1 + v0)] + … + [(T i ) 2 L i + T i (T1a1 + T2a2 + … + T i-1 a i-1 + v0)] = C

[0016] wherein, T i a i is the speed increment corresponding to the i-th cubic Bézier curve, a i is the acceleration of the planned road segment corresponding to the i-th cubic Bézier curve, T iis the planned time for the i-th cubic Bezier curve, C is the lateral offset between the initial position and the desired end point of the host vehicle, L i is the increment coefficient of the planned section corresponding to the i-th cubic Bezier curve.

[0017] In some embodiments, the acceleration a of the planned section corresponding to the i-th cubic Bezier curve i and the increment coefficient L of the planned section corresponding to the i-th cubic Bezier curve i are described by the following formula:

[0018]

[0019] where P 0i is the first control point of the i-th cubic Bezier curve, P 1i is the second control point of the i-th cubic Bezier curve, P 2i is the third control point of the i-th cubic Bezier curve, P 3i is the fourth control point of the i-th cubic Bezier curve.

[0020] In some embodiments, the following method is used to calculate the lateral offset between the initial position of the host vehicle and the desired end point:

[0021] Determine the lateral offset between the initial position of the host vehicle and the desired end point according to the first lateral offset and the second lateral offset;

[0022] The first lateral offset is the lateral offset between the center of the rear axle of the host vehicle at the initial position and the first lane line in the avoidance direction; the second lateral offset is the lateral offset between the center of the rear axle of the host vehicle at the desired end point and the second lane line in the avoidance direction.

[0023] In some embodiments, determining whether to avoid an adjacent vehicle according to the lateral relative distance between the host vehicle and the adjacent vehicle and the calibrated safety distance threshold, or according to the lateral relative speed between the host vehicle and the adjacent vehicle and the calibrated collision time threshold includes:

[0024] If the lateral relative distance between the host vehicle and the adjacent vehicle is less than the calibrated safety distance threshold, it is determined that the path planning condition is satisfied, otherwise it is determined that the path planning condition is not satisfied;

[0025] Alternatively, if the ratio of the lateral relative distance to the lateral relative speed between the host vehicle and the adjacent vehicle is less than the calibrated collision time threshold, it is determined that the path planning condition is satisfied, otherwise the path planning condition is not satisfied.

[0026] In some embodiments, the following method is used to calibrate the safety distance threshold and the collision time threshold:

[0027] Based on a preset scenario, a simulation software is used to simulate the driving processes of the host vehicle and the adjacent vehicle, and the lane width, the set function sensitivity, and the relative speed between the host vehicle and the adjacent vehicle are collected to obtain a data set;

[0028] Based on the data set, a deep learning algorithm is used for prediction to obtain the safety distance threshold and the collision time threshold.

[0029] In some embodiments, the multi-segment cubic Bézier curve is a three-segment cubic Bézier curve.

[0030] In a second aspect, an embodiment of the present application provides a terminal device, which includes a processor and a memory. The memory stores a computer program, and the processor is configured to execute the computer program to implement an implementation method for avoiding an adjacent vehicle provided in the first aspect of the present application.

[0031] In a third aspect, an embodiment of the present application provides a vehicle, which includes a vehicle body, a front-view camera module, an ultrasonic radar module, and a control module;

[0032] The front-view camera module is configured to identify the vehicle in front of the host vehicle and the adjacent vehicles on the first left lane, the second left lane, the first right lane, and the second right lane on the left side of the host vehicle;

[0033] The ultrasonic radar module is configured to detect the adjacent vehicles on the left front side, the right front side, the left side, the right side, the left rear side, and the right rear side of the host vehicle;

[0034] The control module is configured to implement an implementation method for avoiding an adjacent vehicle provided in the first aspect of the present application to avoid the adjacent vehicle.

[0035] The embodiments of the present application have the following beneficial effects:

[0036] In this application, it is determined whether to avoid an adjacent vehicle based on the lateral relative distance between the host vehicle and the adjacent vehicle and the calibrated safety distance threshold, or based on the lateral relative speed between the host vehicle and the adjacent vehicle and the calibrated collision time threshold. If it is determined to avoid the adjacent vehicle, all control points corresponding to multiple cubic Bézier curves are calculated based on the lateral speed at the planned desired end point and the speed increment of each planned section, as well as based on the lateral offset between the initial position of obstacle avoidance and the desired end point and the distance increment of each planned section. Then, multiple cubic Bézier curves are obtained based on all the control points. Among them, each cubic Bézier curve corresponds to a planned section. The corresponding planned section is calculated based on each cubic Bézier curve, and then the host vehicle is controlled to travel according to each planned section to avoid the adjacent vehicle. In this application, it is first determined whether the adjacent vehicle will collide with the host vehicle based on two of the lateral relative distance, lateral relative speed, calibrated safety distance threshold, and collision time threshold. If it is determined that the adjacent vehicle will collide with the host vehicle, then the adjacent vehicle is avoided. A multi-segment cubic Bézier curve is used to plan the avoidance path of the adjacent vehicle. The calculation speed is fast, the avoidance path can be quickly planned, and the continuity of the acceleration is ensured, ensuring that the change of the steering wheel is continuous and does not mutate, and the vehicle control is smoother. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the accompanying drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0038] Figure 1 Fig. shows a schematic structural diagram of a vehicle according to an embodiment of the present application;

[0039] Figure 2 Fig. shows another schematic structural diagram of a vehicle according to an embodiment of the present application;

[0040] Figure 3 Fig. shows a flowchart of a method for realizing the avoidance of an adjacent vehicle according to an embodiment of the present application;

[0041] Figure 4 Fig. shows a schematic diagram of a scenario involved in the method for realizing the avoidance of an adjacent vehicle according to an embodiment of the present application;

[0042] Figure 5 Fig. shows another schematic diagram of a scenario involved in the method for realizing the avoidance of an adjacent vehicle according to an embodiment of the present application.

[0043] MAIN SYMBOL DESCRIPTION OF COMPONENTS

[0044] 100 - Own vehicle; 110 - Vehicle body; 120 - Front view camera module; 121 - Front view camera; 130 - Ultrasonic radar module; 131 - Ultrasonic radar; 140 - Control module; 200 - Adjacent vehicle. Detailed implementation manner

[0045] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0046] The components of the embodiments of the present application usually described and illustrated in the accompanying drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application claimed, but only represents the selected embodiments of the present application. All other embodiments obtained by those skilled in the art based on the embodiments of the present application without creative efforts belong to the scope of protection of the present application.

[0047] In the following, the terms "including", "having" and their cognates that can be used in various embodiments of the present application are only intended to represent specific features, numbers, steps, operations, elements, components or combinations of the foregoing items, and should not be construed as first excluding the existence of one or more other features, numbers, steps, operations, elements, components or combinations of the foregoing items or increasing the possibility of one or more features, numbers, steps, operations, elements, components or combinations of the foregoing items. In addition, the terms "first", "second", "third", etc. are only used for differential description and cannot be understood as indicating or implying relative importance.

[0048] Unless otherwise defined, all terms (including technical terms and scientific terms) used here have the same meaning as commonly understood by those of ordinary skill in the art to which the various embodiments of the present application belong. The terms (such as those defined in a commonly used dictionary) will be interpreted as having the same meaning as the contextual meaning in the relevant technical field and will not be interpreted as having an idealized meaning or being overly formal, unless clearly defined in the various embodiments of the present application.

[0049] The following will describe some implementation manners of the present application in detail with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0050] AEB (Autonomous Emergency Braking / Automatic Emergency Braking, Automatic Emergency Braking System) is an active safety technology whose core function is to sense the environment, analyze risks and actively trigger braking to reduce the probability of collision accidents or mitigate the degree of accident injuries.

[0051] LDP: (Lane Departure Prevention, Lane Departure Prevention System) is a driving assistance function that actively controls the steering wheel to intervene in the lateral movement of the vehicle, aiming to reduce the risk of unconscious lane departure caused by driver distraction.

[0052] ELK (Emergency Lane Keeping Assist, Emergency Lane Keeping Assist System) is an active safety function designed to reduce the risk of the vehicle unconsciously deviating from the lane or colliding with adjacent vehicles through steering intervention, especially in high-speed scenarios.

[0053] In order to prevent a vehicle traveling in the same direction as the host vehicle in the adjacent lane from crashing into the host vehicle, the present application provides a method, a vehicle and a terminal device for avoiding an adjacent vehicle.

[0054] The present application first provides a vehicle. Exemplarily, the vehicle (corresponding to the host vehicle 100) includes a vehicle body 110, a front view camera module 120, an ultrasonic radar module 130 and a control module 140, as Figure 1 shown.

[0055] The front view camera module 120 is used to identify the vehicles in front of the host vehicle 100 and the adjacent vehicles 200 in the first left lane, the second left lane, the first right lane and the second right lane on the left side of the host vehicle 100.

[0056] The ultrasonic radar module 130 is used to detect the adjacent vehicles 200 on the left front side, the right front side, the left side, the right side, the left rear side and the right rear side of the host vehicle 100.

[0057] The control module 140 is used to implement the method for realizing the avoidance of the adjacent vehicle 200 in the present application to avoid the adjacent vehicle 200.

[0058] Exemplarily, as Figure 2 shown, the front view camera module 120 includes a front view camera 121, and the ultrasonic radar module 130 includes 6 ultrasonic radars 131; the front view camera 121 is arranged at the middle position of the front end of the vehicle; the 6 ultrasonic radars 131 are evenly distributed on both sides of the vehicle.

[0059] The main idea of this application is as follows: By adopting the planning algorithm of the three-segment and three-order Bessel curve, a smooth obstacle avoidance path can be planned quickly and effectively. This application uses the lateral relative speed between the adjacent vehicle 200 approaching the host vehicle 100 and the host vehicle 100 as the planning parameter, and at the same time uses the time to collision and the distance to collision as the trigger points to plan a safe obstacle avoidance path.

[0060] The following will illustrate the implementation method for avoiding the adjacent vehicle in combination with some specific embodiments.

[0061] Figure 3 Fig. shows a flowchart of an implementation method for avoiding the adjacent vehicle according to an embodiment of this application. Exemplarily, the implementation method for avoiding the adjacent vehicle includes the following steps:

[0062] S100, determine whether to avoid the adjacent vehicle 200 according to the lateral relative distance between the host vehicle 100 and the adjacent vehicle 200 and the calibrated safety distance threshold, or according to the lateral relative speed between the host vehicle 100 and the adjacent vehicle 200 and the calibrated time to collision threshold.

[0063] In one implementation manner, determining whether to avoid the adjacent vehicle 200 according to the lateral relative distance between the host vehicle 100 and the adjacent vehicle 200 and the calibrated safety distance threshold, or according to the lateral relative speed between the host vehicle 100 and the adjacent vehicle 200 and the calibrated time to collision threshold includes:

[0064] If the lateral relative distance between the host vehicle 100 and the adjacent vehicle 200 is less than the calibrated safety distance threshold, it is determined that the path planning condition is satisfied; otherwise, it is determined that the path planning condition is not satisfied. That is, as Figure 4 shown, when the adjacent vehicle 200 approaches the host vehicle 100, it is determined that the obstacle avoidance condition is satisfied when the following formula is met:

[0065] y_relative_distance<distan

[0066] where, y_relative _ distance is the lateral relative distance between the host vehicle 100 and the adjacent vehicle 200, and distan is the calibrated safety distance threshold.

[0067] Or, if the ratio of the lateral relative distance between the host vehicle 100 and the adjacent vehicle 200 to the lateral relative speed is less than the calibrated time to collision threshold, it is determined that the path planning condition is satisfied; otherwise, the path planning condition is not satisfied. That is, or when the adjacent vehicle 200 approaches the host vehicle 100, it is determined that the obstacle avoidance condition is satisfied when the following formula is met:

[0068] y_ttc<ttc

[0069]

[0070] Among them, y_relative_speed is the relative lateral speed of the host vehicle 100 and the adjacent vehicle 200, and ttc is the calibrated collision time threshold.

[0071] In one implementation, in order to reduce the difficulty and danger of on-vehicle calibration and enable rapid calibration, the following method is used to calibrate the safety distance threshold and the collision time threshold:

[0072] Based on a preset scenario, a simulation software is used to simulate the driving processes of the host vehicle 100 and the adjacent vehicle 200, and the lane width, the set function sensitivity, and the relative speed between the host vehicle 100 and the adjacent vehicle 200 are collected to obtain a data set; the sensitivity is the speed of the host vehicle to avoid obstacles. A high sensitivity means a fast obstacle avoidance speed, and a low sensitivity means a slow obstacle avoidance speed.

[0073] Based on the data set, a deep learning algorithm is used for prediction to obtain the safety distance threshold and the collision time threshold.

[0074] Exemplarily, through the CarSim simulation software for simulation, relevant scenarios are constructed for data collection, and then the safety distance threshold distan is output through a deep learning algorithm.

[0075] By using CarSim to batch construct ideal trigger obstacle avoidance scenarios, data such as the relative speed between the host vehicle 100 and the accident vehicle (the adjacent vehicle 200 crashing into the host vehicle 100), the lane width, and the function sensitivity settings are collected. Then these data are input into a deep learning network for learning to obtain the safety distance threshold distan. Through this calibration method, the danger of on-vehicle calibration can be effectively reduced while the actual obstacle avoidance performance can be improved.

[0076] S200, if it is determined to avoid the adjacent vehicle 200, based on the lateral speed at the expected end point planned, the speed increment of each planned section, and based on the lateral offset between the initial position of obstacle avoidance and the expected end point and the distance increment of each planned section, all control points corresponding to multiple cubic Bézier curves are calculated, and then multiple cubic Bézier curves are obtained based on all control points; among them, each cubic Bézier curve corresponds to a planned section.

[0077] In other words, if it is determined to avoid the adjacent vehicle 200, then based on the preset first calculation formula and second calculation formula, all control points corresponding to multiple cubic Bézier curves are calculated, and then multiple cubic Bézier curves are obtained based on all control points; among them, the first calculation formula is determined according to the lateral speed at the expected end point planned and the speed increment of each planned section; each cubic Bézier curve corresponds to a planned section; the second calculation formula is determined according to the lateral offset between the initial position of obstacle avoidance and the expected end point and the distance increment of each planned section. Among them, the initial position of obstacle avoidance is the position where obstacle avoidance starts.

[0078] When it is determined that the obstacle avoidance condition is met, record the lateral relative distance between the current moment, the host vehicle 100 and the adjacent vehicle 200, and the relative lateral speed between the host vehicle 100 and the adjacent vehicle 200. The adjacent vehicle 200 is a vehicle in the adjacent lane.

[0079] Further, based on the lateral speed at the expected end point of the planned route, the speed increment of each planned route segment, and based on the lateral offset between the initial position of obstacle avoidance and the expected end point and the distance increment of each planned route segment, all control points corresponding to multiple third-order Bezier curves are calculated, including:

[0080] Based on the lateral speed at the expected end point of the planned route, based on the planned time and acceleration of each planned route segment, and based on the lateral offset between the initial position of obstacle avoidance and the expected end point and the planned time, initial lateral speed and increment coefficient of each planned route segment, all control points corresponding to multiple third-order Bezier curves are calculated; wherein, the acceleration is determined according to the control points of the corresponding third-order Bezier curve; the increment coefficient is determined according to the control points of the corresponding third-order Bezier curve.

[0081] In other words, the speed increment of each planned route segment is determined according to the planned time and acceleration of each planned route segment; wherein, the acceleration is determined according to the control points of the corresponding third-order Bezier curve;

[0082] The distance increment of each planned route segment is determined according to the planned time, initial lateral speed and increment coefficient of each planned route segment; the increment coefficient is determined according to the control points of the corresponding third-order Bezier curve.

[0083] In one implementation, the planned times of multiple third-order Bezier curves are set as T1, T2,..., T i ; i is the total number of planned route segments.

[0084] Based on the lateral speed at the expected end point of the planned route, based on the planned time and acceleration of each planned route segment, and based on the lateral offset between the initial position of obstacle avoidance and the expected end point and the planned time, initial lateral speed and increment coefficient of each planned route segment, all control points corresponding to multiple third-order Bezier curves are calculated, including calculating all control points using the following formula:

[0085] T1a1 + T2a2 +... + T i a i = 0 (one)

[0086] (T1 2 L1 + T1v0) + [(T2) 2 L2 + T2(T1a1 + v0)] +... + [(T i ) 2 L i + Ti (T1a1 + T2a2 + … + T i-1 a i-1 + v0)] = C (Two)

[0087] Wherein, T i a i is the speed increment corresponding to the i-th cubic Bézier curve, a i is the acceleration of the planned road section corresponding to the i-th cubic Bézier curve, T i is the planned time of the i-th cubic Bézier curve, C is the lateral offset between the initial position of the host vehicle 100 and the desired end point, L i is the increment coefficient of the planned road section corresponding to the i-th cubic Bézier curve.

[0088] Exemplarily, the multiple cubic Bézier curves are three cubic Bézier curves. The planned times of the three cubic Bézier curves are set as T1, T2, and T3 respectively. The planned times of the three cubic Bézier curves are obtained through calibration. This planned time can affect the riding experience of passengers. The shorter the time, the sharper the planned path; the longer the time, the smoother the planned path. Each section is roughly in the range of 0.7s to 1.5s. This application is calibrated according to the sensitivity of the function settings. The higher the sensitivity of the function settings, the smaller T, the sharper the trajectory, and the faster the obstacle avoidance.

[0089] According to the preset first calculation formula and second calculation formula, all control points corresponding to the three cubic Bézier curves are calculated, including calculating all control points using the following formula:

[0090] T1a1 + T2a2 + T3a3 = 0

[0091] (T1 2 L1 + T1v0) + [(T2) 2 L2 + T2(T1a1 + v0)] + [(T2) 2 L3 + T3(T1a1 + T2a2 + v0)] = C

[0092] Wherein, it is known that the lateral speed at the desired end point is 0, T i a i is the speed increment corresponding to the i-th cubic Bézier curve, a i is the acceleration of the planned road section corresponding to the i-th cubic Bézier curve, T i is the planned time of the i-th cubic Bézier curve, C is the lateral offset between the initial position of the host vehicle 100 and the desired end point, L i is the increment coefficient of the planned road section corresponding to the i-th cubic Bézier curve.

[0093] According to the acceleration-displacement formula, in formula (Two), (T1 2(L1 + T1v0) is the distance increment of the planned section corresponding to the first cubic Bézier curve, (T2) 2 (L2 + T2(T1a1 + v0)) is the distance increment of the planned section corresponding to the second cubic Bézier curve, (T) 2 (L3 + T3(T1a1 + T2a2 + v0)) is the distance increment of the planned section corresponding to the third cubic Bézier curve; the sum of the three is the length of the entire planned path.

[0094] Further, the acceleration a of the planned section corresponding to the i-th cubic Bézier curve i and the increment coefficient L of the planned section corresponding to the i-th cubic Bézier curve i are described by the following formula:

[0095]

[0096] where P 0i is the first control point of the i-th cubic Bézier curve, P 1i is the second control point of the i-th cubic Bézier curve, P 2i is the third control point of the i-th cubic Bézier curve, P 3i is the fourth control point of the i-th cubic Bézier curve.

[0097] The control points of the three cubic Bézier curves are the following data: [k0, k0, 0, 0] [0, 0, k1, k1] [k1, k1, 0, 0]. Substitute each set of control points into formula (III) and formula (IV) respectively, and then substitute a i , L i into formula (I) and formula (II) correspondingly. According to formula (I) and formula (II), k0 and k1 can be calculated, and then twelve control points can be obtained. Substitute the twelve control points into the Bézier curve, and the equations of the three cubic Bézier curves can be obtained.

[0098] Further, the acceleration a of the planned section corresponding to the i-th cubic Bézier curve i is determined according to the formula ΔV = T2 * V(s) corresponding to the change increment of speed and s = 1. s is the position ratio of the point on the Bézier curve. The increment coefficient L of the planned section corresponding to the i-th cubic Bézier curve i is determined according to the formula ΔV = T2 * V(s), s = 1 and the planned time.

[0099] In one implementation, the following method is used to calculate the lateral offset between the initial position of the vehicle 100 and the desired end point:

[0100] Determine the lateral offset between the initial position of the host vehicle 100 and the desired end point according to the first lateral offset and the second lateral offset; wherein, the first lateral offset is the lateral offset between the center of the rear axle of the host vehicle 100 at the initial position and the first lane line in the avoidance direction; the second lateral offset is the lateral offset between the center of the rear axle of the host vehicle 100 at the desired end point and the second lane line in the avoidance direction.

[0101] Exemplarily, the following formula is used to calculate the lateral offset C between the initial position of the host vehicle 100 and the desired end point:

[0102]

[0103] Wherein, is the first lateral offset, is the second lateral offset.

[0104] The formula (V) is obtained in the following manner:

[0105] In the L2-level driving assistance system, the perception module transmits lane line information to the control module 140 for route planning, for example, transmitting the coefficients of the cubic curve equation. Wherein, the expression of the cubic curve equation is:

[0106] y = c0 + c1x + c2x 2 + c3x 3 (VI)

[0107] Wherein, y represents the lateral offset between the center of the rear axle of the vehicle and the lane line when the longitudinal direction is x, c0 represents the lateral offset between the center of the rear axle of the vehicle at the origin and the lane line, c1 is the heading angle of the lane line, c2 is the road curvature, c3 is the road curvature change rate, and x is the longitudinal coordinate value.

[0108] According to the formula (VI), the cubic curve equations corresponding to the first lane line Y_1 and the second lane line Y_2 in the avoidance direction can be expressed as:

[0109]

[0110] According to the formula (VII) and the formula (VIII), the calculation formula for the lateral offset between the current position (initial position) and the desired end point can be expressed as:

[0111]

[0112] S300. Calculate the corresponding planned sections based on each cubic Bézier curve segment, and then control the host vehicle 100 to travel according to each planned section to avoid the adjacent vehicle 200.

[0113] A cubic Bézier curve has four control points, namely the starting point P0, the ending point P3, and two intermediate control points P1 and P2. The starting point and the ending point are the starting position and the final arrival position of the vehicle's travel, while the intermediate control points are used to control the shape and direction of the curve to avoid obstacles or meet other travel requirements. In step S200, the control points of each segment of the cubic Bézier curve are calculated, that is, 12 control points, such as [k0, k0, 0, 0][0, 0, k1, k1][k1, k1, 0, 0].

[0114] For each planned road segment, the expression of the cubic Bézier curve is:

[0115] B(t) = (1 - t) 3 P0 + 3t(1 - t) 2 P1 + 3t 2 (1 - t)P2 + t 3 P3, t ∈ [0, 1]. Where B(t) represents the points on the curve, where P0, P1, P2, P3 are known, t is a parameter. By changing the value of t, different points on the curve can be obtained, thus depicting the entire curve, which is the corresponding planned road segment, and the vehicle is controlled to travel according to this planned road segment. Specifically, take values within the range of [0, 1] at a certain step size △t, substitute t into the cubic Bézier curve equation, and calculate a series of path points B(t n ), n = 0, 1, 2,..., N, and these path points form the route for the host vehicle 100 to travel. The smaller the step size △t, the denser the path points and the more accurate the route.

[0116] This application uses multiple segments of cubic Bézier curves, which can ensure the continuity of acceleration, ensure that the change of the steering wheel is continuous and will not mutate, and the vehicle control is smoother.

[0117] This application uses a cubic Bézier curve to represent the change of the acceleration of the steering wheel, ensuring the continuity of the acceleration of the acceleration of the acceleration of the steering wheel (derived twice). In the prior art, only the continuity of the acceleration of the steering wheel acceleration (derived once) is considered. Therefore, using a cubic Bézier curve in this application is already accurate enough. Using fourth-order, fifth-order, and sixth-order curves will only increase the calculation amount and cost, without optimizing the control effect.

[0118] Exemplarily, in the embodiments of this application, the multiple segments of cubic Bézier curves are three segments of cubic Bézier curves.

[0119] Next, taking the example of planning a path according to three segments of cubic Bézier curves to avoid the adjacent vehicle 200 to the left lane, the reasoning process of this application is introduced as follows, as Figure 5 shown:

[0120] S210, receive the coefficients of the cubic curve equation transmitted by the perception module in the L2-level driving assistance system. The coefficients include the lateral offset c0 between the center of the vehicle's rear axle and the lane line at the origin, the heading angle c1 of the lane line, the road curvature c2, and the road curvature change rate c3.

[0121] The expression of the cubic curve equation is:

[0122] y = c0 + c1x + c2x 2 + c3x 3 (1)

[0123] Among them, y represents the lateral offset between the center of the vehicle's rear axle and the lane line when the longitudinal direction is x, c0 represents the lateral offset between the center of the vehicle's rear axle and the lane line at the origin, c1 is the heading angle of the lane line, c2 is the road curvature, c3 is the road curvature change rate, and x is the longitudinal coordinate value.

[0124] According to equation (1), the cubic curve equation Y_ leftleft (corresponding to Y_2) of the second lane line on the left side of the current vehicle 100 and the cubic curve equation Y_ left (corresponding to Y_1) of the first lane line on the left side of the current vehicle 100 can be expressed as:

[0125]

[0126] Therefore, the lateral offset between the current position and the target point can be calculated as:

[0127]

[0128] In equation (4), C is the lateral offset between the current position of the vehicle 100 and the expected end point.

[0129] S220, calculate the expected end point.

[0130] The expected end point determined by the lateral offset C between the current position of the vehicle 100 and the expected end point and a preset distance.

[0131] S230, calculate the acceleration corresponding to the planned path.

[0132] Determine a third-order Bezier curve planning time T i , so the acceleration a of the planned path can be calculated path , as shown in the following equation (5):

[0133]

[0134] Among them, a path is equal to ΔV in equations (14) and (15) below.

[0135] S240, Bessel curve acceleration formula.

[0136] A(s) = P0(1 - s) 3 + 3P1s(1 - s) 2 + 3P2s 2 (1 - t) + P3s 3 (6)

[0137] In the above formula, P0, P1, P2, and P3 are the control points of the cubic Bessel curve, s ∈ [0, 1], and s is the position ratio of the point on the Bessel curve.

[0138] S250, Bessel curve velocity formula.

[0139] Integrating equation (6) gives the Bessel curve velocity formula:

[0140]

[0141] Expanding equation (7) gives:

[0142]

[0143] S260, Bessel curve position formula.

[0144] Integrating equation (8) gives the Bessel curve position formula:

[0145]

[0146] Expanding equation (9):

[0147]

[0148] S270, Solving for the control points of the Bessel curve.

[0149] Divide a cubic Bessel curve into n equal parts, so the time for each part is:

[0150]

[0151] According to the definition of s, we have:

[0152]

[0153] According to the definition of acceleration, the formula can be deduced:

[0154]

[0155] According to equation (13), the velocity increment can be deduced:

[0156] ΔV = T * V(s) (14)

[0157] Solve the equations simultaneously, and then substitute \(s = 1\) into the equations. This gives the end point of this section of the Bezier curve. Solving for the unknowns based on the boundary conditions gives:

[0158]

[0159] Similarly, the position increment can be obtained as:

[0160]

[0161] That is, formula (16) is derived using the same method as formula (15).

[0162] The above are the derivation formulas for a cubic Bezier curve. In this application, three sections of cubic Bezier curves are used as the obstacle avoidance path (planned path). Each section of the cubic Bezier curve has 4 control points. Based on the obstacle avoidance path, these 12 control points can be initially set as:

[0163] [k0,k0,0,0][0,0,k1,k1][k1,k1,0,0]

[0164] That is, the control points P0, P1, P2, and P3 of the first section of the cubic Bezier curve are [k0,k0,0,0]. By analogy, the control points of the second and third sections of the cubic Bezier curves are [0,0,k1,k1] and [k1,k1,0,0] respectively.

[0165] The planning time for the three sections of cubic Bezier curves is set as: T, 2T, T. In other words, T1 = T, T2 = 2T, and T3 = T.

[0166] Based on formula (15) and the lateral acceleration of the expected planned end point being 0, we can obtain:

[0167] T*a1 + 2Ta2 + T*a3 = 0 (17)

[0168] Based on formula (15) and the set initial position, the initial lateral velocity y_relative_speed of the adjacent vehicle 200 (the accident vehicle hitting the self-vehicle 100) relative to the self-vehicle 100, and the lateral offset C of the expected end point, we can obtain:

[0169] (T 2 *L1 + T*v0) + (2T) 2 *L2 + 2T(T*a1 + v0)] + ((T) 2 *L3 + T(T*a1 + 2T*a2 + v0)) = C (18)

[0170] In formula (17) and formula (18),

[0171] Based on a i 、Li By combining equations (17) and (18), k0 and k1 can be obtained. By combining the 12 control points, the values of the 12 control points of the three-segment third-order Bezier curves can be obtained, and the path planning can be completed to avoid the neighboring car 200 from hitting the own car 100.

[0172] The three-segment third-order Bezier curves in this application can effectively balance path smoothness, computational efficiency, and security.

[0173] The present application also provides an implementation device for avoiding neighboring vehicles in an embodiment. Exemplarily, the implementation device for avoiding neighboring vehicles includes: a judgment module, a control point calculation module and a control processing module.

[0174] A judgment module, used to determine whether to avoid the neighboring vehicle 200 according to the lateral relative distance between the vehicle 100 and the neighboring vehicle 200 and a calibrated safety distance threshold, or according to the lateral relative speed between the vehicle 100 and the neighboring vehicle 200 and a calibrated collision time threshold;

[0175] A control point calculation module is used to calculate all control points corresponding to multiple third-order Bezier curves based on the lateral speed at the planned expected end point and the speed increment of each planned section, and the lateral offset between the initial position of the obstacle avoidance and the expected end point and the distance increment of each planned section, if the obstacle avoidance neighboring vehicle 200 is determined, and then obtain multiple third-order Bezier curves based on all the control points; wherein each third-order Bezier curve corresponds to a planned section;

[0176] The control processing module is used to calculate the corresponding planned road section based on each third-order Bezier curve, and then control the vehicle to move according to each planned road section to avoid the neighboring vehicle 200.

[0177] It can be understood that the device of this embodiment corresponds to the implementation method of avoiding neighboring vehicles in the above embodiment, and the optional items in the above embodiment are also applicable to this embodiment, so they will not be repeated here.

[0178] The present application also provides a terminal device, which exemplary comprises a processor and a memory, wherein the memory stores a computer program, and the processor runs the computer program, so that the terminal device executes the functions of each module in the above-mentioned method for avoiding neighboring vehicles or the above-mentioned device for avoiding neighboring vehicles. Exemplarily, the terminal device is a controller of a vehicle.

[0179] Among them, the processor can be an integrated circuit chip with signal processing capabilities. The processor can be a general-purpose processor, including at least one of a central processing unit (CPU), a graphics processing unit (GPU), a network processor (NP), a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc., and can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application.

[0180] The memory can be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. Among them, the memory is used to store a computer program, and after receiving an execution instruction, the processor can execute the computer program accordingly.

[0181] The present application also provides a computer-readable storage medium for storing the computer program used in the above terminal device. For example, the computer-readable storage medium can include, but is not limited to: various media such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disc that can store program code.

[0182] In several embodiments provided by this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and structural diagrams in the accompanying drawings show the possible architectures, functions, and operations of devices, methods, and computer program products according to multiple embodiments of this application. In this regard, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in an alternative implementation, the functions marked in the blocks can occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks can actually be executed substantially in parallel, and they can sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the structural diagram and / or flowchart, as well as the combination of blocks in the structural diagram and / or flowchart, can be implemented by a dedicated hardware-based system that executes the specified functions or actions, or can be implemented by a combination of dedicated hardware and computer instructions.

[0183] In addition, in each embodiment of this application, the various functional modules or units can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part.

[0184] If the described functions are implemented in the form of software function modules and sold or used as an independent product, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a smart phone, a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of this application.

[0185] As described above, the above are only the specific implementation manners of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, and all should be covered by the protection scope of this application.

Claims

1. An implementation method for avoiding adjacent vehicles, characterized in that, Including: Determine whether to avoid the adjacent vehicle according to the lateral relative distance between the host vehicle and the adjacent vehicle and the calibrated safety distance threshold, or according to the lateral relative speed between the host vehicle and the adjacent vehicle and the calibrated collision time threshold; If it is determined to avoid the adjacent vehicle, calculate all control points corresponding to multiple cubic Bézier curves based on the lateral speed at the planned desired end point and the speed increment of each planned section, and based on the lateral offset between the initial position of obstacle avoidance and the desired end point and the distance increment of each planned section. Then, obtain multiple cubic Bézier curves based on all the control points; wherein, each cubic Bézier curve corresponds to a planned section; Calculate the corresponding planned section based on each cubic Bézier curve respectively, and then control the host vehicle to travel according to each planned section to avoid the adjacent vehicle.

2. The implementation method for avoiding adjacent vehicles according to claim 1, wherein The calculating all control points corresponding to multiple cubic Bézier curves based on the lateral speed at the planned desired end point and the speed increment of each planned section, and based on the lateral offset between the initial position of obstacle avoidance and the desired end point and the distance increment of each planned section includes: Calculate all control points corresponding to multiple cubic Bézier curves based on the lateral speed at the planned desired end point, the planned time and acceleration of each planned section, and based on the lateral offset between the initial position of obstacle avoidance and the desired end point and the planned time, initial lateral speed and increment coefficient of each planned section; wherein, the acceleration is determined according to the control points of the corresponding cubic Bézier curve; the increment coefficient is determined according to the control points of the corresponding cubic Bézier curve.

3. The implementation method for avoiding adjacent vehicles according to claim 2, wherein Set the planning times of the multi-segment third-order Bézier curves to be T1, T2, …, T i ; i is the total number of planned road segments; The calculating all control points corresponding to multiple cubic Bézier curves based on the lateral speed at the planned desired end point, the planned time and acceleration of each planned section, and based on the lateral offset between the initial position of obstacle avoidance and the desired end point and the planned time, initial lateral speed and increment coefficient of each planned section includes calculating all control points using the following formula: T1a1 + T2a2 + … + T i a i = 0 (T1 2 L1 + T1v0) + [(T2) 2 L2 + T2(T1a1 + v0)] + … + [(T i ) 2 L i + T i (T1a1 + T2a2 + … + T i-1 a i-1 + v0)] = C Among them, T i a i is the speed increment corresponding to the third-order Bezier curve of the i-th segment, and a i is the acceleration of the planned road section corresponding to the third-order Bezier curve of the i-th segment. T i is the planned time of the third-order Bezier curve of the i-th segment. C is the lateral offset between the initial position of the host vehicle and the desired end point, and L i is the increment coefficient of the planned road section corresponding to the third-order Bezier curve of the i-th segment.

4. The implementation method for avoiding adjacent vehicles according to claim 3, wherein The acceleration a of the planned road section corresponding to the i-th cubic Bézier curve i and the increment coefficient L of the planned road section corresponding to the i-th cubic Bézier curve i are described by the following formula: Among them, P 0i is the first control point of the third-order Bezier curve of the i-th segment, P 1i is the second control point of the third-order Bezier curve of the i-th segment, P 2i is the third control point of the third-order Bezier curve of the i-th segment, P 3i is the fourth control point of the third-order Bezier curve of the i-th segment.

5. The implementation method for avoiding adjacent vehicles according to claim 1, characterized in that Adopt the following method to calculate the lateral offset between the initial position of the host vehicle and the desired end point: Determine the lateral offset between the initial position of the host vehicle and the desired end point according to the first lateral offset and the second lateral offset; The first lateral offset is the lateral offset between the center of the rear axle of the host vehicle at the initial position and the first lane line in the avoidance direction; the second lateral offset is the lateral offset between the center of the rear axle of the host vehicle at the desired end point and the second lane line in the avoidance direction.

6. The implementation method for avoiding adjacent vehicles according to claim 1, characterized in that, The determining whether to avoid the adjacent vehicle according to the lateral relative distance between the host vehicle and the adjacent vehicle and the calibrated safety distance threshold, or according to the lateral relative speed between the host vehicle and the adjacent vehicle and the calibrated collision time threshold includes: If the lateral relative distance between the host vehicle and the adjacent vehicle is less than the calibrated safety distance threshold, it is determined that the path planning condition is satisfied; otherwise, it is determined that the path planning condition is not satisfied; Or, if the ratio of the lateral relative distance between the host vehicle and the adjacent vehicle to the lateral relative speed is less than the calibrated collision time threshold, it is determined that the path planning condition is satisfied; otherwise, the path planning condition is not satisfied.

7. The implementation method for avoiding adjacent vehicles according to claim 6, characterized in that, The safety distance threshold and the collision time threshold are calibrated by the following method: Based on a preset scenario, a simulation software is used to simulate the driving processes of the host vehicle and the adjacent vehicle, and the lane width, the set functional sensitivity, and the relative speed between the host vehicle and the adjacent vehicle are collected to obtain a data set; Based on the data set, a deep learning algorithm is used for prediction to obtain the safety distance threshold and the collision time threshold.

8. The method for realizing avoiding adjacent vehicles according to any one of claims 1-7, characterized in that, The multi-segment cubic Bézier curve is a three-segment cubic Bézier curve.

9. A terminal device, characterized in that, The terminal device includes a processor and a memory. The memory stores a computer program, and the processor is configured to execute the computer program to implement the method for avoiding an adjacent vehicle according to any one of claims 1-8.

10. A vehicle, characterized in that, The vehicle includes a vehicle body, a front view camera module, an ultrasonic radar module, and a control module; The front view camera module is configured to identify the vehicle in front of the host vehicle and the adjacent vehicles on the first left lane, the second left lane, the first right lane, and the second right lane of the host vehicle; The ultrasonic radar module is configured to detect the adjacent vehicles on the left front side, the right front side, the left side, the right side, the left rear side, and the right rear side of the host vehicle; The control module is configured to implement the method for avoiding an adjacent vehicle according to any one of claims 1-8 to avoid an adjacent vehicle.

Images