CN-116605238-B - Target vehicle cut-in intention prediction method and device

CN116605238BCN 116605238 BCN116605238 BCN 116605238BCN-116605238-B

Abstract

The invention relates to a target vehicle cut-in intention prediction method and device, which comprise the steps of obtaining different motion states of a target vehicle in different scenes, and predicting the cut-in intention of the target vehicle according to the different motion states by adopting different judgment standards, so that the problem that the cut-in intention of the target vehicle cannot be accurately predicted in the prior art is solved.

Inventors

LIU YIDI
Xiao Kaixing
YAN XU
YANG DONGFANG

Assignees

重庆长安汽车股份有限公司

Dates

Publication Date: 20260505
Application Date: 20230621

Claims (14)

1. A target vehicle cut-in intention prediction method, characterized by comprising: acquiring different motion states of a target vehicle in different scenes; According to the different motion states, different judgment standards are adopted to predict the cutting intention of the target vehicle; wherein, the predicting the cutting intention of the target vehicle according to the different motion states by adopting different determination standards includes: Predicting the cutting-in intention of the target vehicle according to the transverse distance between the target vehicle and the lane where the automatic driving vehicle is located and the longitudinal distance between the target vehicle and the automatic driving vehicle, wherein the lane comprises a lane central line or a lane boundary line, or predicting the cutting-in intention of the target vehicle according to the transverse distance between the target vehicle and the lane where the automatic driving vehicle is located; The predicting the cutting intention of the target vehicle according to the transverse distance between the target vehicle and the lane where the automatic driving vehicle is located and the longitudinal distance between the target vehicle and the automatic driving vehicle comprises: acquiring a first lateral distance between the target vehicle and the lane; judging whether the first transverse distance is smaller than a first distance judging parameter; if the first lateral distance is smaller than the first distance judgment parameter, calculating the product of the first lateral distance and the first lateral speed of the target vehicle to obtain a first prediction result; if the first prediction result is smaller than zero, calculating a first distance from the target vehicle to the automatic driving vehicle; if the first distance is smaller than or equal to a set first distance threshold, acquiring a first longitudinal distance between the target vehicle and the automatic driving vehicle; And if the first longitudinal distance is larger than a set longitudinal distance threshold, determining that the target vehicle has a cutting-in intention.
2. The method of claim 1, wherein the different motion states of the target vehicle include any one of a target vehicle out of lane state, a target vehicle into lane state, a target vehicle line, a target vehicle traveling near a lane where an autonomous vehicle is traveling and not entering a lane where the autonomous vehicle is traveling.
3. The prediction method according to claim 1, wherein either the predicting of the intention of cut of the target vehicle based on a lateral distance between the target vehicle and a lane in which an autonomous vehicle is located and a longitudinal distance between the target vehicle and the autonomous vehicle includes: Judging whether the lane where the target vehicle is located and the lane where the automatic driving vehicle is located have intersection or not; if the intersection exists, calculating a second distance from the target vehicle to the automatic driving vehicle; if the second distance is smaller than or equal to a set second distance threshold, acquiring a second longitudinal distance between the target vehicle and the automatic driving vehicle; if the second longitudinal distance is greater than a set longitudinal distance threshold, acquiring a second transverse distance between the target vehicle and a lane where the automatic driving vehicle is located; if the second lateral distance is smaller than a second distance judgment parameter, calculating the product of the second lateral distance and the second lateral speed of the target vehicle to obtain a second prediction result; if the second prediction result is smaller than zero, judging whether the second transverse speed of the target vehicle is larger than a set second transverse speed threshold; If yes, determining that the target vehicle has the cutting-in intention.
4. The prediction method according to claim 1, wherein either the performing of the cut-in intention prediction on the target vehicle according to the lateral distance between the target vehicle and the lane in which the autonomous vehicle is located includes: Judging whether the target vehicle is in line pressing or not; if yes, calculating the product of the third transverse distance of the target vehicle and the third transverse speed of the target vehicle to obtain a third prediction result; if the third prediction result is smaller than zero, judging whether the line pressing parameter is larger than a line pressing parameter judging threshold value, wherein the line pressing parameter comprises line pressing quantity and line pressing quantity change rate; if yes, acquiring a third transverse speed of the target vehicle; If the third transverse speed is greater than the speed judgment parameter, determining that the target vehicle has a cutting-in intention; And if the third transverse speed is smaller than the speed judging parameter, determining whether the target vehicle has a cutting-in intention or not according to the line pressing quantity and the course angle of the target vehicle.
5. A prediction method according to claim 1 or 3, characterized in that the distance determination parameter is calculated by: Calculating a course angle of the driving direction of the target vehicle relative to the driving direction of the lane where the automatic driving vehicle is located; Acquiring the length and width of the target vehicle, the width of a lane where the target vehicle is located, the first transverse speed of the target vehicle and the transverse time interval; And calculating the distance judging parameter according to the course angle, the vehicle length, the vehicle width, the lane width, the first transverse speed and the transverse time interval, or calculating the distance judging parameter according to the course angle, the vehicle length, the vehicle width, the first transverse speed and the transverse time interval.
6. The prediction method according to claim 1, 3 or 4, characterized in that the method further comprises: acquiring a fourth transverse distance between the target vehicle and the lane and a course angle of the running direction of the target vehicle relative to the running direction of the lane where the automatic driving vehicle is located; if the fourth transverse distance is not different from the heading angle, judging whether the target vehicle is in line pressing or not; If no line is pressed, calculating the product of the fourth transverse distance and the fourth transverse speed of the target vehicle to obtain a fourth prediction result, or judging whether the fourth transverse speed is less than or equal to a second transverse speed threshold; And if the fourth prediction result is greater than or equal to zero, or the fourth transverse speed is less than or equal to a third transverse speed threshold, determining that the target vehicle has no cutting-in intention.
7. The method of predicting as recited in claim 6, further comprising: acquiring a fifth lateral distance between the target vehicle and the lane; judging whether the fifth transverse distance is larger than a first vehicle track width threshold value; If the width of the lane is larger than the width threshold, calculating the product of the fifth transverse distance and the fifth transverse speed of the target vehicle to obtain a fifth prediction result, or judging whether the fifth transverse speed is smaller than or equal to a fifth transverse speed threshold; If the fifth prediction result is smaller than zero or the fifth transverse speed is smaller than a fifth transverse speed threshold, calculating the product of a fifth transverse distance between the target vehicle and the lane and a course angle of the running direction of the target vehicle relative to the running direction of the lane where the automatic driving vehicle is located, and obtaining a sixth prediction result; If the sixth prediction result is smaller than zero or the course angle is smaller than a first angle threshold value, calculating a third distance between the target vehicle and a lane where the automatic driving vehicle is located; If the third distance is smaller than a distance threshold parameter or the fifth transverse distance is smaller than a second road width threshold, acquiring a longitudinal time interval of the target vehicle; and if the longitudinal time interval is within the set range, determining that the target vehicle belongs to the intermediate state of the cutting intention.
8. A target vehicle cut-in intention prediction apparatus, comprising: The motion state module is used for acquiring different motion states of the target vehicle in different scenes; the prediction module is used for predicting the cutting intention of the target vehicle according to the different motion states by adopting different judgment standards; wherein the prediction module comprises: A first prediction intention unit, configured to predict a cut intention of the target vehicle according to a lateral distance between the target vehicle and a lane where the autonomous vehicle is located and a longitudinal distance between the target vehicle and the autonomous vehicle, where the lane includes a lane center line or a lane boundary line, or predict the cut intention of the target vehicle according to the lateral distance between the target vehicle and the lane where the autonomous vehicle is located; Wherein the first prediction intention unit includes: A first lateral sub-module for acquiring a first lateral distance between the target vehicle and the lane; The first judging submodule is used for judging whether the first transverse distance is smaller than a first distance judging parameter or not; The first prediction submodule is used for calculating the product of the first transverse distance and the first transverse speed of the target vehicle if the first distance is smaller than the first distance judgment parameter to obtain a first prediction result; a first calculation sub-module, configured to calculate a first distance from the target vehicle to an autonomous vehicle if the first prediction result is less than zero; The first longitudinal sub-module is used for acquiring a first longitudinal distance between the target vehicle and the automatic driving vehicle if the first distance is smaller than or equal to a set first distance threshold value; And the first cutting-in submodule is used for determining that the target vehicle has cutting-in intention if the first longitudinal distance is larger than a set longitudinal distance threshold value.
9. The prediction apparatus according to claim 8, wherein the movement state of the target vehicle includes any one of a target vehicle out of lane state, a target vehicle merge into lane state, a target vehicle line-up, a target vehicle traveling near a lane where an autonomous vehicle is located, and not entering a lane where the autonomous vehicle is located.
10. The prediction apparatus according to claim 8, wherein either the first prediction intention unit includes: The second judging submodule is used for judging whether the lane where the target vehicle is located and the lane where the automatic driving vehicle is located have intersection or not; A second calculation sub-module for calculating a second distance from the target vehicle to the autonomous vehicle if there is an intersection; The second longitudinal sub-module is used for acquiring a second longitudinal distance between the target vehicle and the automatic driving vehicle if the second distance is smaller than or equal to a set second distance threshold value; The second transverse sub-module is used for acquiring a second transverse distance between the target vehicle and the lane where the automatic driving vehicle is located if the second longitudinal distance is larger than a set longitudinal distance threshold; the second prediction submodule is used for calculating the product of the second transverse distance and the second transverse speed of the target vehicle to obtain a second prediction result if the second transverse distance is smaller than a second distance judgment parameter; a third judging sub-module, configured to judge whether a second lateral speed of the target vehicle is greater than a set second lateral speed threshold if the second prediction result is less than zero; and the second cutting-in sub-module is used for determining that the target vehicle has cutting-in intention if the second cutting-in sub-module is larger than the second transverse speed threshold value.
11. The prediction apparatus according to claim 8, wherein either the first prediction intention unit includes: A fourth judging sub-module, configured to judge whether the target vehicle is in line; A third calculation sub-module, configured to calculate, if yes, a product of a third lateral distance of the target vehicle and a third lateral speed of the target vehicle, to obtain a third prediction result; A fifth judging sub-module, configured to judge whether a crimping parameter is greater than a crimping parameter judging threshold if the third prediction result is less than zero, where the crimping parameter includes crimping amount and crimping amount change rate; The third transverse sub-module is used for acquiring a third transverse speed of the target vehicle if yes; And the third cutting-in sub-module is used for determining whether the target vehicle has a cutting-in intention or not according to the line pressing quantity and the course angle of the target vehicle if the third transverse speed is smaller than the speed judging parameter.
12. The prediction apparatus according to claim 8 or 10, wherein the distance determination parameter is calculated by: Calculating a course angle of the driving direction of the target vehicle relative to the driving direction of the lane where the automatic driving vehicle is located; Acquiring the length and width of the target vehicle, the width of a lane where the target vehicle is located, the first transverse speed of the target vehicle and the transverse time interval; And calculating the distance judging parameter according to the course angle, the vehicle length, the vehicle width, the lane width, the first transverse speed and the transverse time interval, or calculating the distance judging parameter according to the course angle, the vehicle length, the vehicle width, the first transverse speed and the transverse time interval.
13. The prediction device according to claim 8, 10 or 11, characterized in that the device further comprises: the first acquisition module is used for acquiring a fourth transverse distance between the target vehicle and the lane and a course angle of the running direction of the target vehicle relative to the running direction of the lane where the automatic driving vehicle is located; The first judging module is used for judging whether the target vehicle is in line pressing or not if the fourth transverse distance is different from the heading angle; the first calculation module is used for obtaining a fourth prediction result according to the product of the fourth transverse distance and the fourth transverse speed of the target vehicle if no line is pressed, or judging whether the fourth transverse speed is smaller than or equal to a second transverse speed threshold; And the output module is used for determining that the target vehicle does not have the cutting-in intention if the fourth prediction result is greater than or equal to zero or the fourth transverse speed is less than or equal to a third transverse speed threshold.
14. The prediction device according to claim 13, characterized in that the device further comprises: a second acquisition module for acquiring a fifth lateral distance between the target vehicle and the lane; the second judging module is used for judging whether the fifth transverse distance is larger than a first vehicle width threshold value or not; The second calculation module is used for calculating the product of the fifth transverse distance and the fifth transverse speed of the target vehicle to obtain a fifth prediction result if the second transverse speed is larger than the lane width threshold value, or judging whether the fifth transverse speed is smaller than or equal to a fifth transverse speed threshold value; The deviation module is used for calculating the product of the fifth transverse distance between the target vehicle and the lane and the course angle of the running direction of the target vehicle relative to the running direction of the lane where the automatic driving vehicle is located if the fifth prediction result is smaller than zero or the fifth transverse speed is smaller than a fifth transverse speed threshold value, so as to obtain a sixth prediction result; The central line distance module is used for calculating a third distance between the target vehicle and the lane where the automatic driving vehicle is located if the sixth prediction result is smaller than zero or the course angle is smaller than a first angle threshold value; The longitudinal time interval module is used for acquiring the longitudinal time interval of the target vehicle if the third distance is smaller than a distance threshold parameter or the fifth transverse distance is smaller than a second vehicle channel width threshold; and the state module is used for determining that the target vehicle belongs to the intermediate state of the cutting-in intention if the longitudinal time interval is within a set range.

Description

Target vehicle cut-in intention prediction method and device Technical Field The invention relates to the technical field of intelligent safe driving, in particular to a target vehicle cut-in intention prediction method and device. Background With the continuous development of automatic driving and assisted driving technologies, automobiles are gradually entering the intelligent era. However, how to accurately predict the cut-in channel behavior (commonly called "stopover") of a target vehicle, so as to ensure the safety and the comfort of automatic driving is still a worth exploring problem. Aiming at the problem of target vehicle cut-in intention prediction, the main current method for predicting the target vehicle cut-in intention adopts a neural network model to predict the target vehicle cut-in intention, and the neural network model can predict the target vehicle cut-in intention in a typical application scene with obvious movement trend for the target vehicle, however, a plurality of atypical cut-in scenes exist in a real road, such as driving style is aggressive, cut-in behaviors which do not follow traffic rules or scenes with complex intersection road relations, and training samples are less because atypical cut-in scenes are not always present, so the neural network model cannot accurately predict the target vehicle cut-in intention, and the automatic driving safety is affected. Disclosure of Invention The present invention provides a method for predicting a cutting-in intention of a target vehicle to solve the problem that the cutting-in intention of the target vehicle cannot be accurately predicted in the prior art, and a device for predicting the cutting-in intention of the target vehicle. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: A target vehicle plunge intention prediction method comprises the steps of obtaining different motion states of a target vehicle in different scenes, and predicting the plunge intention of the target vehicle according to the different motion states by adopting different judgment standards. According to the technical means, different motion states of the target vehicle in different scenes are obtained, and according to the different motion states of the target vehicle, the cutting-in intention prediction is carried out on the target vehicle by adopting different judgment standards, so that the prediction of various cutting-in intentions of the vehicle can be more rapidly and accurately realized without collecting training samples, and the safety of automatic driving is improved. Further, the movement state of the target vehicle includes any one of a target vehicle out of lane state, a target vehicle merge into lane state, a target vehicle line pressing, a target vehicle traveling near a lane where an autonomous vehicle is located and not entering the lane where the autonomous vehicle is located. According to the technical means, the motion state of the vehicle can be determined. Further, the predicting the intention of the target vehicle according to the different motion states by adopting different determination criteria includes: And predicting the cutting-in intention of the target vehicle according to the transverse distance between the target vehicle and the lane where the automatic driving vehicle is located and the longitudinal distance between the target vehicle and the automatic driving vehicle, wherein the lane comprises a lane central line or a lane boundary line, or predicting the cutting-in intention of the target vehicle according to the transverse distance between the target vehicle and the lane where the automatic driving vehicle is located. According to the technical means, the cutting-in intention prediction is performed on the target vehicle according to the distance between the target vehicle and the lane where the automatic driving vehicle is located, so that the problem that the cutting-in intention of the target vehicle cannot be accurately predicted due to the fact that training samples are few and the safety of automatic driving is improved can be avoided. Further, the predicting the cut-in intention of the target vehicle according to the lateral distance between the target vehicle and the lane in which the autonomous vehicle is located and the longitudinal distance between the target vehicle and the autonomous vehicle includes: Acquiring a first lateral distance P d between the target vehicle and the lane; Judging whether the first transverse distance is smaller than a first distance judgment parameter D offset; if the first lateral distance P d is smaller than the first distance judging parameter, calculating the product of the first lateral distance P d and the first lateral speed V d of the target vehicle to obtain a first prediction result; if the first prediction result is smaller than zero, calculating a first distance from the target vehicle to the automatic driving vehicle; If