A method for identifying and controlling the cut-in and cut-out of the target ahead during adaptive cruise control

Abstract

The present invention discloses a method for identifying and controlling the cut-in and cut-out of a target ahead of an adaptive cruise, comprising the following steps: identifying, collecting, and fusing information of targets and lane lines on a road through an on-board sensor; simulating the driver's focus on the road and the target while driving through an element extraction algorithm, and extracting the road and target focus elements; setting the element extraction boundary according to the physical characteristics of the vehicle and the driver's general operating habits, and numerically describing the elements; combining various elements for preliminary identification to obtain the target motion intention; combining the target motion intention, unifying the numerical description as a discrimination condition, and realizing the motion risk assessment of the targets around the vehicle; and formulating the acceleration and deceleration strategy for the vehicle according to the motion risk assessment to control the acceleration and deceleration of the vehicle. The present invention adopts a decision-making mechanism that simulates the driver's discrimination to optimize the recognition of the scene, and has the advantages of intuitive algorithm, small amount of calculation, good versatility and practicality.

Description

Technical Field

[0001] The invention belongs to the field of adaptive cruise, and in particular relates to a method for identifying and controlling cutting-in and cutting-out of a front target in an adaptive cruise. Background Art

[0002] Adaptive cruise control is a component of the driving assistance system. It automatically controls the time distance when following the vehicle in front and the speed when cruising based on the perception of the vehicle in front. It is widely used on highways, urban elevated roads and non-extreme road conditions. While this function brings convenience to users, there are also some scenarios where the function cannot be effectively covered or the performance is poor. Among them, timely selecting the entry target in the entry scene to accelerate and decelerate, and timely releasing the original target in the exit scene are the most complained by users during use, and are also the scenes that engineering and technical personnel are constantly optimizing. Summary of the invention

[0003] The purpose of the present invention is to provide a method for identifying and controlling the cut-in and cut-out of a target ahead in an adaptive cruise control, which adopts a decision-making mechanism that simulates driver discrimination to optimize the scene recognition, and has the advantages of intuitive algorithm, small amount of computation, good versatility and practicality.

[0004] In order to solve the above technical problems, the technical solution of the present invention is: an adaptive cruise forward target cut-in and cut-out recognition and control method, comprising the following steps:

[0005] Identify, collect and fuse information of targets and lane lines on the road through on-board sensors; the collected information includes at least target position, target motion state information and lane line position;

[0006] The feature extraction algorithm is used to simulate the driver's focus on the road and the target while driving, and the road and target focus elements are extracted;

[0007] Set the feature extraction boundary according to the physical characteristics of the vehicle and the general operating habits of the driver, and describe the features numerically;

[0008] Combine various elements to perform preliminary identification to obtain the target's movement intention;

[0009] Combined with the target motion intention, the numerical description is unified as the discrimination condition to achieve the motion risk assessment of the targets around the vehicle;

[0010] According to the motion risk assessment, the acceleration and deceleration strategy of the ego vehicle is formulated to control the acceleration and deceleration of the ego vehicle;

[0011] The target motion state information specifically includes: target lateral position, target longitudinal position, target lateral speed, target longitudinal speed, target lateral acceleration, target longitudinal acceleration, target yaw angle, target width, target turn signal, and target lateral position standard deviation, target longitudinal position standard deviation, target lateral speed standard deviation, target longitudinal speed standard deviation, target lateral acceleration standard deviation, and target longitudinal acceleration standard deviation established during the tracking process.

[0012] The specific process of lane line position recognition and collection is as follows:

[0013] The lane lines are segmented from the road background and fitted. The results of lane line recognition serve as auxiliary conditions for subsequent target motion risk assessment and the basis for road right division. The lane line fitting method is:

[0014] y=c0+c1x+c2x 2 +c3x 3

[0015] Among them, x is the longitudinal distance of the lane line description point relative to the vehicle, y is the lateral distance of the lane line description point relative to the vehicle, c0, c1, c2, and c3 are the first fitting coefficient, the second fitting coefficient, the third fitting coefficient, and the fourth fitting coefficient, respectively.

[0016] The extracted elements include the target turn signal, target lateral speed, current road curvature radius R, target effective yaw angle The lane to which the target contour center point C belongs, the width distribution percentage of the lane to which the target belongs, the longitudinal pursuit time Δt between the vehicle and the target, and the overlap rate iou between the target and the vehicle in the width direction width .

[0017] The current road curvature radius R is obtained as follows:

[0018] When the lane lines are clear and the recognition quality is high, the road curvature radius R = 1 / (2*c2);

[0019] When there are no lane lines or the lane line recognition quality is not high, the lane curvature radius R is estimated using a dynamic or kinematic model based on the current motion state of the vehicle; the current motion state of the vehicle includes the steering wheel angle, vehicle speed, longitudinal acceleration and yaw rate of the vehicle.

[0020] Target effective yaw angle Measuring yaw angle for vehicle-mounted sensors The angle between the tangent line of the lane line at the target location and the longitudinal axis of the vehicle The difference between the lane line tangent line at the target location and the longitudinal axis of the vehicle measured by the on-board sensor is the angle of the target relative to the longitudinal axis of the vehicle.

[0021] The lane to which the center point C of the target contour belongs is the lane where the midpoint C of the line between the leftmost point A and the rightmost point B of the target contour belongs; the lane width distribution percentage of the target belongs to is the ratio of the target width in different lanes to the width of the target vehicle itself. Given that the cut-in and cut-out mainly focus on the lane where the vehicle is located, the left lane of the vehicle, and the right lane of the vehicle, the width distribution percentage set is mainly composed of three parts: the width distribution percentage of the target in this lane WHR e , the target lane width distribution percentage WHR on the left side of the lane l and the target lane width percentage WHR on the right side of the lane r , where the lane is the lane where the ego vehicle is located. Considering that the lane where the ego vehicle is located will change before and after the lane change, and when the ego vehicle actively changes lanes to the adjacent lane (left lane or right lane), it is necessary to consider the lane where the target vehicle is located as a reference to calculate the ego vehicle's cut-in situation, so the above calculation formula is:

[0022] WHR e =WH in_ego / WH obj

[0023] WHR l =max(WH in_left / WH obj , W.H. ego_in_left / WH ego )

[0024] WHR r =max(WH in_right / WH obj , W.H. ego_in_right / WH ego )

[0025] Where WH in_ego The width of the target vehicle entering the lane, WH in_left WH is the width of the left lane of the target vehicle entering the lane. in_right WH is the width of the right lane of the target vehicle entering the lane. obj is the total width of the target vehicle, WH ego_in_left WH is the width of the lane on the left side of the vehicle. ego_in_right WH is the width of the lane on the right side of the vehicle. ego is the total width of the vehicle.

[0026] The longitudinal pursuit time Δt between the ego vehicle and the target is the time required for the ego vehicle and the target to overlap in the longitudinal direction, and its calculation method is:

[0027]

[0028] In the formula, v x is the relative speed between the ego vehicle and the target vehicle in the longitudinal direction, is the longitudinal arc length between the ego vehicle and the target vehicle, and its calculation method is:

[0029]

[0030] Where, Dis OD The radial distance from the center point O of the front bumper of the vehicle along the road direction to the longitudinal position of the target vehicle, the trajectory point D of the vehicle;

[0031] The overlap rate (IOU) between the target and the vehicle in the width direction width , which is calculated as:

[0032] IOU width =0.5*(WH obj +WH ego ) / Dis CD

[0033] Where, Dis CD is the lateral distance between the target contour center point C and the vehicle trajectory point D, WH ego is the vehicle width.

[0034] The specific method of numerically describing the elements is: statistically analyzing the elements and obtaining the value range of each element [Thd min ,Thd max ], analyze and obtain the basic weight coefficient Element of each element for cutting in and out fd , and calibrate the influence coefficient Element within the numerical range of the calculation element coeff , a numerical description of each element Element factor It is expressed as:

[0035] Element factor(A) =Element fd(A) +Element coeff(A) *(A-Thd min(A) )

[0036] In the formula, A is the element being referred to, which is replaced according to the selected element, that is, different elements correspond to different values, and their corresponding relationship is obtained through statistics and calibration; the target turn signal signal is an enumerated element directly recognized by the vehicle sensor, and its corresponding influence coefficient Element coeff is 0.

[0037] According to the extracted elements,

[0038] Select the intended combination of elements, combining the weight influence coefficient Weight and the numerical description of the element Elementfactor , calculate the distribution probability of each intention Motion_probility, the calculation formula is as follows:

[0039] Motion probility =Weight(A)*Element factor(A) +Weight(B)*Element factor(B) +……+Weight(K)*Element factor(K)

[0040] In the formula, the weight influence coefficients of different factors correspond to different values, and their corresponding relationships are obtained through statistics and calibration;

[0041] The relative intentions of the ego vehicle and the target vehicle are determined by comparing the intention distribution probability Motion_probility with the set intention threshold.

[0042] The method of unifying the numerical description into the discrimination condition is: the elements used for risk assessment are digitized and uniformly mapped to the risk assessment threshold IOUT hd_factor In the above, the specific implementation is the product of the numerical description of each risk assessment factor, so as to realize the dynamic adjustment of the threshold value as different factors are superimposed. The specific method is as follows:

[0043] IOU Thd_factor =Element factor(A) *Element factor(B) *……Element factor(N )*IOU Thd

[0044] In the formula, IOU Thd It is the preset basic threshold. Different basic thresholds are used when there is cut-in, cut-out, and no obvious intention to correspond to different response requirements; IOU Thd_factor It is the weight threshold after multiple elements are superimposed;

[0045] The relative motion intention is used as the scene distinction entry, and the overlap rate (IOU) between the target and the vehicle in the width direction is compared in different scenes. width and weight threshold IOU Thd_factor The size of determines whether the target should be selected as the following target in advance, continuously selected as the following target, or released and then other targets are selected. , thereby performing acceleration and deceleration control of the vehicle.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention simulates the risk assessment process of the driver when driving by extracting factors and confirming the initial intention, and unifies the risks corresponding to each factor into the weight threshold IOU.Thd_factor In the process, by comparing the threshold and the set value, the target selection and the final acceleration and deceleration control strategy are realized. It has the advantages of comprehensive consideration of factors, close to real driving habits, high accuracy of target selection, strong operability of the method and good effect experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a flow chart of an embodiment of the present invention;

[0049] Figure 2 This is a schematic diagram of the position of the target vehicle in an embodiment of the present invention;

[0050] Figure 3 It is a schematic diagram of calculating the overlap ratio in the width direction in an embodiment of the present invention;

[0051] Figure 4 It is a schematic diagram of specifying acceleration and deceleration strategies according to motion risk assessment in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0053] The present invention provides a method and device for identifying and controlling cutting-in and cutting-out of a target ahead of an adaptive cruise control. The method comprises: identifying and fusing targets and lane lines on a road by means of a vehicle-mounted sensor; designing an element extraction algorithm by simulating the driver's focus on the road and the target when driving; setting extraction element boundaries according to the physical characteristics of the vehicle and the driver's general operating habits, and numerically describing the elements; preliminarily identifying the target's motion intention in combination with each element; unifying the parameter description into a discrimination condition in combination with the identification intention, and realizing the motion risk assessment of the targets around the vehicle; and performing acceleration and deceleration control on the vehicle based on the risk assessment, so as to improve the safety, comfort and intelligence level of the driving assistance system.

[0054] Through sensors such as cameras, millimeter-wave radars, and lidars, the target position and motion status information on the road are identified, including: lateral position, longitudinal position, lateral speed, longitudinal speed, lateral acceleration, longitudinal acceleration, and the target lateral position standard deviation, longitudinal position standard deviation, lateral speed standard deviation, longitudinal speed standard deviation, lateral acceleration standard deviation, longitudinal acceleration standard deviation, etc. established during the tracking process; as well as target attitude profile attributes, including: target yaw angle, target width, turn signal, etc.

[0055] The vehicle-mounted sensor identifies lane lines on the road, including:

[0056] The lane lines are segmented from the road background by cameras and lidar, and a cubic spline curve is fitted for the lane lines, i.e., y=c0+c1x+c2x 2 +c3x 3 , where x is the longitudinal distance of the lane line description point relative to the vehicle, y is the lateral distance of the lane line description point relative to the vehicle, c0, c1, c2, and c3 are polynomial fitting coefficients respectively. Lane line recognition serves as an auxiliary condition for subsequent target risk assessment and the basis for road right division.

[0057] The main purpose of fusing the target state is to improve the stability of information in each dimension of the target and to improve the accuracy of subsequent extraction of influencing factors.

[0058] The elements include: 1) directly identified enumerated elements: turn signal; 2) directly identified state signal: target lateral speed; 3) calculated elements, mainly including: current road curvature radius R, target effective yaw angle The lane to which the target contour center point C belongs, the width distribution percentage of the lane to which the target belongs, the longitudinal pursuit time Δt between the vehicle and the target, and the overlap rate iou between the target and the vehicle in the width direction width .

[0059] The main functions of the elements are shown in Table 1:

[0060] Table 1

[0061]

[0062]

[0063] The curvature radius R of the current road can be obtained in two ways: one is that when the lane line is clear and the sensor recognition quality is high, the road curvature radius R = 1 / (2*c2); the other is that when there is no lane line or the lane line recognition quality is not high, the current motion state of the vehicle (steering wheel angle, vehicle speed, longitudinal acceleration, yaw rate and other information) is used to estimate the lane curvature radius R using a dynamic or kinematic model. The driving road state estimation is mainly used to set the sensitivity when assessing the risk of target cut-in and cut-out, ensuring that the algorithm can be applied to straight and curved scenes.

[0064] The target effective yaw angle Measuring yaw angle for the sensor The angle between the tangent line of the lane line at the target location and the longitudinal axis of the vehicle Difference: target heading angle measured by the sensor It is the angle between the target and the longitudinal axis of the vehicle, and cannot represent the target's posture intention in the road coordinate system (in order to avoid running out of the lane when the vehicle is driving on the road, the angle between the vehicle body and the tangent of the lane line needs to be kept within a certain range).

[0065] The lane to which the center point C of the target contour belongs is the lane where the midpoint C of the leftmost point A and the rightmost point B of the target contour is located.

[0066] The target lane width distribution percentage is the ratio of the target width in different lanes to the target vehicle's own width. Since the cut-in and cut-out mainly focus on the lane where the vehicle is located, the lane to the left of the vehicle, and the lane to the right of the vehicle, the width distribution percentage set mainly consists of three parts: the target lane width distribution percentage WHR e , the target lane width distribution percentage WHR on the left side of the lane l and the target lane width percentage WHR on the right side of the lane r , where the lane is the lane where the ego vehicle is located. Considering that the lane where the ego vehicle is located will change before and after the lane change, and when the ego vehicle actively changes lanes to the adjacent lane (left lane or right lane), it is necessary to consider the lane where the target vehicle is located as a reference to calculate the ego vehicle's cut-in situation, so the above calculation formula is:

[0067] WHR e =WH in_ego / WH obj

[0068] WHR l =max(WH in_left / WH obj , W.H. ego_in_left / WH ego )

[0069] WHR r =max(WH in_right / WH obj , W.H. ego_in_right / WH ego )

[0070] Where WH in_eg o is the width of the target vehicle entering the lane, WH in_left WH is the width of the left lane of the target vehicle entering the lane. in_right WH is the width of the right lane of the target vehicle entering the lane. obj is the total width of the target vehicle, WH wgo_in_left WH is the width of the lane on the left side of the vehicle. ego_in_right WH is the width of the lane on the right side of the vehicle. egois the total width of the vehicle. The longitudinal pursuit time Δt between the vehicle and the target is the time required for the longitudinal position of the vehicle and the target to overlap, and its calculation method is

[0071]

[0072] In the formula, v x is the relative speed between the ego vehicle and the target vehicle in the longitudinal direction, is the longitudinal arc length between the ego vehicle and the target vehicle, and its calculation method is:

[0073]

[0074] Where R is the same as the curvature radius R of the current road calculated previously, dis OD The radial distance from the center point O of the front bumper of the vehicle along the road direction to the longitudinal position of the target vehicle, the trajectory point D of the vehicle;

[0075] The overlap ratio (IOU) between the target and the vehicle in the width direction width , which is calculated as:

[0076] IOU width =0.5*(WH obj +WH ego ) / Dis CD

[0077] Where, Dis CD is the lateral distance between the target contour center point C and the vehicle trajectory point D, WH ego is the vehicle width.

[0078] The risk assessment factors are numerically described, and the above calculation factors are statistically analyzed to obtain the value range of each factor [Thd min ,Thd max ], analyze and obtain the basic weight coefficient Element of each element for cutting in and out fd , and calibrate the influence coefficient Element within the numerical range of the calculation element coeff , a numerical description of each element Element factor It is expressed as:

[0079] Element factor(A) =Element fd(A) +Element coeff(A) *(A-Thd min(A) )

[0080] In the formula, A is the element being referred to, which is replaced according to the selected element, that is, different elements correspond to different values, and the corresponding relationship is obtained through statistics and calibration.

[0081] For the enumeration state value turn signal influence coefficient Element coeff is 0.

[0082] According to the extracted elements,

[0083] Select the intended combination of elements, combining the weight influence coefficient Weight and the numerical description of the element Element factor , calculate the distribution probability of each intention Motion_probility, the calculation formula is as follows:

[0084] Motion probility =Weight(A)*Element factor(A) +Weight(B)*Elementfa ctor(B) +……+Weight(K)*Element factor( K)

[0085] In the formula, different factor weight influence coefficients Weight correspond to different values, and their corresponding relationships are obtained through statistics and calibration.

[0086] The relative intentions of the ego vehicle and the target vehicle are determined by comparing the intention distribution probability Motion_probility with the set intention threshold.

[0087] The preliminary recognition of the target movement intention is mainly confirmed by the divided intention confirmation elements, and its basic confirmation rules are shown in Table 2:

[0088] Table 2

[0089]

[0090]

[0091]

[0092] The unified discrimination is mainly used to digitize the risk assessment factors and then uniformly map them to the risk assessment threshold IOU. Thd_factor In the above, it is specifically implemented as the product of the numerical description of each risk assessment factor, so as to realize the dynamic adjustment of the threshold as different factors are superimposed, specifically:

[0093] IOU Thd_factor =Element factor(A) *Element factor(B) *......Element factor(N) *IOU Thd

[0094] Where IOU ThdThe basic threshold for setting is different for identifying cut-in, cut-out, and no obvious intention, so as to correspond to different response requirements; IOU Thd_factor It is the weight threshold after multiple elements are superimposed.

[0095] In summary, the calculated elements and directly recognized state information (brake lights, turn signals) are unified to the weight threshold IOU Thd_factor By comparing the overlap rate IOU between the target and the vehicle in the width direction width and weight threshold IOU Thd_factor The size of is used to determine whether the target should be selected as a follow target in advance, continuously selected as a follow target, or released to continue selecting other targets.

[0096] The acceleration and deceleration control of the ego vehicle based on risk assessment mainly involves selecting an acceleration and deceleration control strategy that matches the target according to the current time-distance situation after the target is selected: ensuring that the state fluctuation of the ego vehicle is small in scenarios with lower risks, avoiding misidentification while ensuring the comfort of system operation; and braking in a timely manner in scenarios with higher risks to ensure driving safety.

[0097] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for identifying and controlling the cut-in and cut-out of a forward target in an adaptive cruise control, characterized in that: The following steps are involved: Identify, collect and fuse information of targets and lane lines on the road through on-board sensors; the collected information includes at least target position, target motion state information and lane line position; The feature extraction algorithm is used to simulate the driver's focus on the road and the target while driving, and the road and target focus elements are extracted; Set the feature extraction boundary according to the physical characteristics of the vehicle and the general operating habits of the driver, and describe the features numerically; Combine various elements to perform preliminary identification to obtain the target's movement intention; Combined with the target motion intention, the numerical description is unified as the discrimination condition to achieve the motion risk assessment of the targets around the vehicle; According to the motion risk assessment, the acceleration and deceleration strategy of the ego vehicle is formulated to control the acceleration and deceleration of the ego vehicle; The target motion state information specifically includes: target lateral position, target longitudinal position, target lateral speed, target longitudinal speed, target lateral acceleration, target longitudinal acceleration, target yaw angle, target width, target turn signal, and target lateral position standard deviation, target longitudinal position standard deviation, target lateral speed standard deviation, target longitudinal speed standard deviation, target lateral acceleration standard deviation, target longitudinal acceleration standard deviation established during the tracking process; The extracted elements include the target turn signal, target lateral speed, current road curvature radius R, target effective yaw angle The lane to which the target contour center point C belongs, the width distribution percentage of the lane to which the target belongs, the longitudinal pursuit time Δt between the vehicle and the target, and the overlap rate IOU between the target and the vehicle in the width direction width ; For each extracted element, select the intended element combination, combine the weight influence coefficient Weight and the numerical description of the element Element factor , calculate the distribution probability of each intention Motion_probility, and the calculation formula is: Motion probility =Weight(A)*Element factor(A) +Weight(B)*Element factor(B) +……+Weight(K)*Element factor(K) In the formula, the weight influence coefficients of different factors correspond to different values, and their corresponding relationships are obtained through statistics and calibration; Determine the relative intentions of the ego vehicle and the target vehicle by comparing the intention distribution probability Motion_probility with the set intention threshold; The method of unifying the numerical description as the discrimination condition is: the elements used for risk assessment are digitized and uniformly mapped to the risk assessment threshold IOU Thd_factor In the above, the specific implementation is the product of the numerical description of each risk assessment factor, so as to realize the dynamic adjustment of the threshold value as different factors are superimposed. The specific method is as follows: IOU Thd_factor =Element factor(A) *Element factor(B) *......Element factor(N) *IOU Thd In the formula, IOU Thd It is the preset basic threshold. Different basic thresholds are used when there is cut-in, cut-out, and no obvious intention to correspond to different response requirements; IOU Thd_factor It is the weight threshold after multiple elements are superimposed; The relative motion intention is used as the scene distinction entry, and the overlap rate (IOU) between the target and the vehicle in the width direction is compared in different scenes. width and weight threshold IOU Thd_factor The size of the target determines whether the target should be selected as a follow target in advance, continuously selected as a follow target, or released to continue selecting other targets.

2. The method for identifying and controlling the cut-in and cut-out of the target ahead in adaptive cruise according to claim 1, characterized in that: The specific process of lane line position recognition and collection is as follows: The lane lines are segmented from the road background and fitted. The results of lane line recognition serve as auxiliary conditions for subsequent target motion risk assessment and the basis for road right division. The lane line fitting method is: y=c0+c1x+c2x 2 +c3x 3 Among them, x is the longitudinal distance of the lane line description point relative to the vehicle, y is the lateral distance of the lane line description point relative to the vehicle, c0, c1, c2, and c3 are the first fitting coefficient, the second fitting coefficient, the third fitting coefficient, and the fourth fitting coefficient, respectively.

3. The method for identifying and controlling the cut-in and cut-out of the target ahead in adaptive cruise according to claim 1, characterized in that: The current road curvature radius R is obtained as follows: When the lane lines are clear and the recognition quality is high, the road curvature radius R = 1 / (2*c2); When there are no lane lines or the lane line recognition quality is not high, the lane curvature radius R is estimated using a dynamic or kinematic model based on the current motion state of the vehicle; the current motion state of the vehicle includes the steering wheel angle, vehicle speed, longitudinal acceleration and yaw rate of the vehicle.

4. The method for identifying and controlling the cutting-in and cutting-out of the target ahead in adaptive cruise according to claim 1, characterized in that: Target effective yaw angle Measuring yaw angle for vehicle-mounted sensors The angle between the tangent line of the lane line at the target location and the longitudinal axis of the vehicle difference.

5. The method for identifying and controlling the cut-in and cut-out of the target ahead in adaptive cruise according to claim 1, characterized in that: The lane to which the center point C of the target contour belongs is the lane where the midpoint C of the line between the leftmost point A and the rightmost point B of the target contour belongs; the lane width distribution percentage of the target belongs to is the ratio of the target width in different lanes to the width of the target vehicle itself. Given that the cut-in and cut-out mainly focus on the lane where the vehicle is located, the left lane of the vehicle, and the right lane of the vehicle, its width distribution percentage set is mainly composed of three parts: that is, the width distribution percentage WHR of the target in this lane e , the target lane width distribution percentage WHR on the left side of the lane l and the target lane width percentage WHR on the right side of the lane r , where the lane is the lane where the ego vehicle is located. Considering that the lane where the ego vehicle is located will change before and after the lane change, and when the ego vehicle actively changes lanes to the adjacent lane (left lane or right lane), it is necessary to consider the lane where the target vehicle is located as a reference to calculate the ego vehicle's cut-in situation, so its calculation formula is: WHR e =WH in_ego / WH obj WHR l =max(WH in_left / WH obj ,WH ego_in_left / WH ego ) WHR r =max(WH in_right / WH obj ,WH ego_in_right / WH ego ) Where WH in_ego The width of the target vehicle entering the lane, WH in_left WH is the width of the left lane of the target vehicle entering the lane. in_right WH is the width of the right lane of the target vehicle entering the lane. obj is the total width of the target vehicle, WH ego_in_right WH is the width of the lane on the left side of the vehicle. ego_in_right WH is the width of the lane on the right side of the vehicle. ego is the total width of the vehicle.

6. The method for identifying and controlling the cut-in and cut-out of the target ahead in adaptive cruise according to claim 5, characterized in that: The longitudinal pursuit time Δt between the ego vehicle and the target is the time required for the ego vehicle and the target to overlap in the longitudinal direction, and its calculation method is: In the formula, v x is the relative speed between the ego vehicle and the target vehicle in the longitudinal direction, is the longitudinal arc length between the ego vehicle and the target vehicle, and its calculation method is: Where, Dis OD The radial distance from the center point O of the front bumper of the vehicle to the longitudinal position of the target vehicle along the road direction; The overlap ratio (IOU) between the target and the vehicle in the width direction width , which is calculated as: IOU width =0.5*(WH obj +WH ego ) / Dis CD Where, Dis CD is the lateral distance between the target contour center point C and the vehicle trajectory point D, WH ego is the vehicle width.

7. The method for identifying and controlling the cutting-in and cutting-out of the forward target in adaptive cruise according to claim 1, characterized in that: The specific method of numerically describing the elements is: statistically analyzing the elements and obtaining the value range of each element [Thd min ,Thd max ], analyze and obtain the basic weight coefficient Element of each element for cutting in and out fd , and calibrate the influence coefficient Element within the numerical range of the calculation element coeff , a numerical description of each element Element factor It is expressed as: Element factor(A) =Element fd(A) +Element coeff(A) *(A-Thd min(A) ) In the formula, A is the element being referred to, which is replaced according to the selected element, that is, different elements correspond to different values, and their corresponding relationship is obtained through statistics and calibration; the target turn signal signal is an enumerated element directly recognized by the vehicle sensor, and its corresponding influence coefficient Element coeff is 0.

Images