CN-121989937-A - Intersection-oriented vehicle interaction decision-making method, intersection-oriented vehicle interaction decision-making device, medium, equipment and vehicle

CN121989937ACN 121989937 ACN121989937 ACN 121989937ACN-121989937-A

Abstract

The application discloses a vehicle interactive decision method, a device, a medium, equipment and a vehicle for an intersection, wherein the method comprises the steps of obtaining a system state matrix, a vehicle control matrix and a vehicle control matrix in a linear system state equation of a vehicle and the vehicle, determining a vehicle optimal control gain matrix and a vehicle optimal control gain matrix according to the system state matrix, the vehicle control matrix, the first semi-positive definite matrix, the second semi-positive definite matrix and the second positive definite matrix, and determining a vehicle optimal control quantity at each moment according to the vehicle optimal control gain matrix and the system state quantity at each moment under the condition that the characteristic value of a closed loop system matrix is located in a unit circle, so as to control the vehicle according to the vehicle optimal control quantity at each moment.

Inventors

ZHANG JIAXU
SUN GANG
Teng ting
LI WEI
WANG TENGFEI

Assignees

魔门塔(苏州)科技有限公司

Dates

Publication Date: 20260508
Application Date: 20241101

Claims (10)

1. An intersection-oriented vehicle interactive decision-making method, which is characterized by comprising the following steps: Acquiring a system state matrix, a self-vehicle control matrix and a self-vehicle control matrix in a linear system state equation of a self-vehicle and a self-vehicle, wherein the first half-positive matrix and the first half-positive matrix in a self-vehicle optimizing target in an intersection-oriented scene and the second half-positive matrix in the self-vehicle optimizing target in the intersection-oriented scene, the first half-positive matrix is a coefficient of a system state quantity in the self-vehicle optimizing target, the first half-positive matrix is a coefficient of a self-vehicle control quantity in the self-vehicle optimizing target, the second half-positive matrix is a coefficient of a system state quantity in the self-vehicle optimizing target, and the second half-positive matrix is a coefficient of a self-vehicle control quantity in the self-vehicle optimizing target; determining an own vehicle optimal control gain matrix and an other vehicle optimal control gain matrix according to the system state matrix, the own vehicle control matrix, the other vehicle control matrix, the first half positive definite matrix, the first positive definite matrix, the second half positive definite matrix and the second positive definite matrix, wherein the own vehicle optimal control gain matrix is a feedback Nash equilibrium solution of an own vehicle, and the other vehicle optimal control gain matrix is a feedback Nash equilibrium solution of the other vehicle; And under the condition that the characteristic value of the closed-loop system matrix is located in a unit circle, determining the self-vehicle optimal control quantity at each moment according to the self-vehicle optimal control gain matrix and the system state quantity at each moment so as to control the self-vehicle according to the self-vehicle optimal control quantity at each moment, wherein the closed-loop system matrix is a matrix determined according to the system state matrix, the self-vehicle control matrix, the other-vehicle control matrix, the self-vehicle optimal control gain matrix and the other-vehicle optimal control gain matrix.
2. The method of claim 1, wherein determining whether the eigenvalues of the closed loop system matrix lie within a unit circle comprises: Directly calculating the eigenvalue of the closed loop system matrix, and determining whether the calculated eigenvalue is positioned in a unit circle; or judging whether a preset matrix has observability, if so, determining that the eigenvalue of the closed loop system matrix is positioned in a unit circle, wherein the preset matrix is formed by splicing the system state matrix, the first semi-positive definite matrix and the second semi-positive definite matrix.
3. The method of claim 1, wherein determining a vehicle optimal control gain matrix and a vehicle optimal control gain matrix from the system state matrix, the vehicle control matrix, the first semi-positive definite matrix, the first positive definite matrix, the second semi-positive definite matrix, and the second positive definite matrix comprises: Setting a first matrix and a second matrix in the process of solving the vehicle optimization target and the other vehicle optimization target by using a Pontriya Jin Zuixiao value algorithm, generating an equation about a vehicle optimal control gain matrix, an equation about the other vehicle optimal control gain matrix, a coupling Li-Cared equation about the first matrix and a coupling Li-Cared equation about the second matrix, and solving a target equation set consisting of the equation about the vehicle optimal control gain matrix, the equation about the other vehicle optimal control gain matrix, the coupling Li-Cared equation about the first matrix and the coupling Li-Cared equation about the second matrix to obtain the vehicle optimal control gain matrix and the other vehicle optimal control gain matrix; The first matrix is a matrix for describing a mapping relationship between the cooperative quantity of the self-vehicle discrete Hamiltonian and the system state quantity at the same moment; The second intermediate matrix is a matrix for describing the mapping relationship between the cooperative quantity of the discrete Hamiltonian of the other vehicle and the system state quantity at the same moment; The equation about the own vehicle optimal control gain matrix includes the own vehicle optimal control gain matrix, the first positive definite matrix, the own vehicle control matrix, the first matrix, the system state matrix, the other vehicle control matrix, the other vehicle optimal control gain matrix; the equation about the optimal control gain matrix of the other vehicle comprises the optimal control gain matrix of the other vehicle, the second positive definite matrix, the control matrix of the other vehicle, the second matrix, the system state matrix, the control matrix of the own vehicle and the optimal control gain matrix of the own vehicle; The coupled licark-proposed equation for a first matrix includes the first matrix, the first semi-positive definite matrix, the vehicle-optimum control gain matrix, the first positive definite matrix, the system state matrix, the other vehicle control matrix, the other vehicle-optimum control gain matrix, and the vehicle-optimum control matrix; The coupled licark-proposed equation for the second matrix includes the second matrix, the second semi-positive definite matrix, the other vehicle optimal control gain matrix, the second positive definite matrix, the system state matrix, the other vehicle control matrix, the own vehicle control matrix, and the own vehicle optimal control gain matrix.
4. A method according to claim 3, wherein the set of target equations comprises: Wherein K 1 represents the own vehicle optimal control gain matrix, R 1 represents the first positive definite matrix, B 1 represents the own vehicle control matrix, P 1 represents the first matrix, a represents the system state matrix, B 2 represents the other vehicle control matrix, K 2 represents the other vehicle optimal control gain matrix, R 2 represents the second positive definite matrix, P 2 represents the second matrix, Q 1 represents the first positive definite matrix, and Q 2 represents the second positive definite matrix.
5. The method according to any one of claims 1 to 4, wherein, Wherein x (k) represents a system state quantity at a kth time; When i=1, J 1 represents a vehicle optimization target, Q 1 represents the first positive definite matrix, u 1 (k) represents a vehicle control amount at the kth time, and R 1 represents the first positive definite matrix; When i=2, J 2 represents a vehicle optimization target, Q 2 represents the second semi-positive definite matrix, u 2 (k) represents the other vehicle control amount at the kth time, and R 2 represents the second positive definite matrix.
6. An intersection-oriented vehicle interactive decision device, the device comprising: An acquisition unit configured to acquire a system state matrix, a vehicle control matrix, and a vehicle control matrix in a linear system state equation for a vehicle and a vehicle, a first semi-positive definite matrix and a first positive definite matrix in a vehicle optimization target in an intersection-oriented scene, and a second semi-positive definite matrix and a second positive definite matrix in a vehicle optimization target in an intersection-oriented scene, wherein the first semi-positive definite matrix is a coefficient of a system state quantity in the vehicle optimization target, the first positive definite matrix is a coefficient of a vehicle control quantity in the vehicle optimization target, the second semi-positive definite matrix is a coefficient of a system state quantity in the vehicle optimization target, and the second positive definite matrix is a coefficient of a vehicle control quantity in the vehicle optimization target; The first determining unit is configured to determine an optimal control gain matrix of the own vehicle and an optimal control gain matrix of the other vehicle according to the system state matrix, the control matrix of the own vehicle, the control matrix of the other vehicle, the first half positive definite matrix, the first positive definite matrix, the second half positive definite matrix and the second positive definite matrix, where the optimal control gain matrix of the own vehicle is a feedback nash equilibrium solution of the own vehicle, and the optimal control gain matrix of the other vehicle is a feedback nash equilibrium solution of the other vehicle; And the second determining unit is used for determining the optimal control quantity of the own vehicle at each moment according to the optimal control gain matrix of the own vehicle and the system state quantity at each moment under the condition that the characteristic value of the closed-loop system matrix is located in a unit circle, so as to control the own vehicle according to the optimal control quantity of the own vehicle at each moment, wherein the closed-loop system matrix is a matrix determined according to the system state matrix, the control matrix of the own vehicle, the control matrix of the other vehicle, the optimal control gain matrix of the own vehicle and the optimal control gain matrix of the other vehicle.
7. The apparatus of claim 6, wherein the apparatus further comprises: The judging unit is used for determining whether the characteristic value of the closed-loop system matrix is positioned in the unit circle; The judging unit is specifically configured to directly calculate a eigenvalue of the closed-loop system matrix, and determine whether the calculated eigenvalue is located in a unit circle, or judge whether a preset matrix has observability, if the preset matrix has observability, determine that the eigenvalue of the closed-loop system matrix is located in the unit circle, where the preset matrix is formed by splicing the system state matrix, the first semi-positive definite matrix and the second semi-positive definite matrix.
8. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, implements the method according to any of claims 1-5.
9. An electronic device, the electronic device comprising: One or more processors; The processor is coupled with a storage device for storing one or more programs; The one or more programs, when executed by the one or more processors, cause the electronic device to implement the method of any of claims 1-5.
10. A vehicle comprising the apparatus of any one of claims 6-7 or the electronic device of claim 9.

Description

Intersection-oriented vehicle interaction decision-making method, intersection-oriented vehicle interaction decision-making device, medium, equipment and vehicle Technical Field The application relates to the technical field of intelligent driving, in particular to a vehicle interactive decision method, device, medium, equipment and vehicle for an intersection. Background In the crossing traffic interaction scene of the own vehicle and other vehicles, a reasonable interaction decision strategy is an effective means for improving the crossing vehicle traffic efficiency. In order to improve the traffic efficiency of the crossing, a joint optimization mode of the own vehicle and the other vehicle is generally adopted, and the joint optimization weights of the own vehicle and the other vehicle are manually configured according to experience. However, this manual weight configuration method based on experience is poor in flexibility, and cannot cope with the temporary self-strategy of the other vehicles, so that the traffic efficiency of the crossing vehicles is reduced. Disclosure of Invention The application provides a vehicle interactive decision method, a device, a medium, equipment and vehicles for an intersection, which can improve the traffic efficiency of vehicles at the intersection. The specific technical scheme is as follows: In a first aspect, an embodiment of the present application provides a vehicle interactive decision method for an intersection, where the method includes: Acquiring a system state matrix, a self-vehicle control matrix and a self-vehicle control matrix in a linear system state equation of a self-vehicle and a self-vehicle, wherein the first half-positive matrix and the first half-positive matrix in a self-vehicle optimizing target in an intersection-oriented scene and the second half-positive matrix in the self-vehicle optimizing target in the intersection-oriented scene, the first half-positive matrix is a coefficient of a system state quantity in the self-vehicle optimizing target, the first half-positive matrix is a coefficient of a self-vehicle control quantity in the self-vehicle optimizing target, the second half-positive matrix is a coefficient of a system state quantity in the self-vehicle optimizing target, and the second half-positive matrix is a coefficient of a self-vehicle control quantity in the self-vehicle optimizing target; determining an own vehicle optimal control gain matrix and an other vehicle optimal control gain matrix according to the system state matrix, the own vehicle control matrix, the other vehicle control matrix, the first half positive definite matrix, the first positive definite matrix, the second half positive definite matrix and the second positive definite matrix, wherein the own vehicle optimal control gain matrix is a feedback Nash equilibrium solution of an own vehicle, and the other vehicle optimal control gain matrix is a feedback Nash equilibrium solution of the other vehicle; And under the condition that the characteristic value of the closed-loop system matrix is located in a unit circle, determining the self-vehicle optimal control quantity at each moment according to the self-vehicle optimal control gain matrix and the system state quantity at each moment so as to control the self-vehicle according to the self-vehicle optimal control quantity at each moment, wherein the closed-loop system matrix is a matrix determined according to the system state matrix, the self-vehicle control matrix, the other-vehicle control matrix, the self-vehicle optimal control gain matrix and the other-vehicle optimal control gain matrix. According to the scheme, the embodiment of the application can firstly determine the optimal control gain matrix of the own vehicle and the optimal control gain matrix of the other vehicle according to the system state matrix, the own vehicle control matrix and the other vehicle control matrix in the linear system state equation of the own vehicle and the other vehicle, the first semi-positive fixed matrix and the first positive fixed matrix in the own vehicle optimization target in the crossing-oriented scene, and the second semi-positive fixed matrix and the second positive fixed matrix in the other vehicle optimization target in the crossing-oriented scene, and can determine the optimal control quantity of the own vehicle at each moment according to the optimal control gain matrix of the own vehicle and the system state quantity at each moment under the condition that the characteristic value of the closed loop system matrix is determined to be positioned in a unit circle, so as to control the own vehicle according to the optimal control quantity of the own vehicle at each moment. The self-vehicle optimal control gain matrix and the other vehicle optimal control gain matrix are feedback Nash equilibrium solutions, and the characteristic values of the closed loop system matrix are located in a unit circle, so that the sel