CN-121982932-A - Vehicle scheduling method and device, vehicle and storage medium

CN121982932ACN 121982932 ACN121982932 ACN 121982932ACN-121982932-A

Abstract

The application relates to the technical field of vehicle control and discloses a vehicle scheduling method, a device, a vehicle and a storage medium, wherein the method comprises the steps of acquiring running state information from a vehicle to a position before an intersection without a traffic signal lamp, and acquiring relative positions of other vehicles in the current intersection area and predicted running path information; the method comprises the steps of determining a corresponding multi-objective optimization function based on a traffic interaction scene of a vehicle in each sampling period, determining the traffic interaction scene based on running state information of the vehicle and relative positions and predicted running path information of other vehicles, constructing a longitudinal dynamics model of the vehicle based on the running state information of the vehicle, combining the multi-objective optimization function and safety constraint conditions to construct a model prediction control problem, solving the model prediction control problem, obtaining a target control input sequence in a control time domain, and outputting the first control input quantity in the target control input sequence to an execution mechanism of the vehicle, so that real-time regulation and control of the vehicle are realized.

Inventors

LI CAIWEI
JIANG ZIMING
FU RENTAO
KONG DEBAO

Assignees

中国第一汽车股份有限公司

Dates

Publication Date: 20260505
Application Date: 20260202

Claims (10)

1. A vehicle scheduling method, characterized by comprising: acquiring running state information before a self-vehicle runs to an intersection without traffic lights, and acquiring relative positions of other vehicles in the current intersection area and predicted running path information; In each sampling period, determining a corresponding multi-objective optimization function based on a traffic interaction scene where a vehicle is located, wherein the traffic interaction scene is determined based on the running state information of the vehicle and the relative positions and predicted running path information of other vehicles; Constructing a longitudinal dynamics model of the vehicle based on the running state information of the vehicle, and constructing a model predictive control problem by combining the multi-objective optimization function and the safety constraint condition; Solving the model predictive control problem to obtain a target control input sequence in a control time domain, and outputting the first control input quantity in the target control input sequence to an execution mechanism of the own vehicle to realize real-time regulation and control of the own vehicle.
2. The vehicle scheduling method according to claim 1, characterized in that the constructing a vehicle longitudinal dynamics model based on the running state information of the own vehicle includes: Based on a nonlinear continuous dynamic equation of longitudinal movement of the vehicle, carrying out local linearization processing at the initial state of the current sampling period to obtain a linearized state space expression; discretizing the state space expression to obtain a discrete time state space model; And combining the ground rolling friction resistance item into a control input variable to form an equivalent control input quantity, and assuming that a system state transition matrix and the control input matrix are kept unchanged in a prediction time domain so as to construct the longitudinal dynamics model of the vehicle, wherein displacement and speed are used as state variables, and the equivalent control input quantity is used as a control variable.
3. The vehicle scheduling method according to claim 1, wherein constructing a model predictive control problem based on the vehicle longitudinal dynamics model in combination with the multi-objective optimization function, and the safety constraint condition, includes: deriving a predicted state sequence of the own vehicle in a prediction time domain based on the longitudinal dynamics model; Converting the safety constraint into a set of inequality constraints with respect to the state prediction sequence and the control sequence; And constructing the model predictive control problem based on the multi-objective optimization function and the inequality constraint set.
4. The vehicle scheduling method of claim 1, wherein the traffic interaction scenario comprises a collision-free scenario, a linear collision scenario, a punctual collision scenario, and a compound collision scenario: the collision-free scene is a risk scene in which potential collision does not exist between the own vehicle and other vehicles; the linear conflict scene is a linear conflict scene in which the self-vehicle and other vehicles on the same entrance lane have rear-end collision risks; the punctiform conflict scene is a punctiform conflict scene that the self-vehicle and other vehicles from different entrance lanes have paths crossing in the crossing; The composite conflict scene is a composite conflict scene containing both linear conflict risks and punctiform conflict risks.
5. The vehicle scheduling method of claim 4, wherein the determining a corresponding multi-objective optimization function based on the traffic interaction scenario comprises: when the self-vehicle is in the collision-free scene, a first multi-objective optimization function with traffic efficiency index, energy economy index and riding comfort index as optimization sub-objectives is adopted; When the vehicle is in the linear conflict scene, a second-type multi-objective optimization function which takes a line safety performance index as a main optimization sub-objective and takes the traffic efficiency index, the energy economy index and the riding comfort index as auxiliary optimization sub-objectives is adopted; When the vehicle is in the punctiform conflict scene, a third type multi-objective optimization function which takes a conflict point safety performance index as a main optimization sub-objective and takes the traffic efficiency index, the energy economy index and the riding comfort index as auxiliary optimization sub-objectives is adopted; When the self-vehicle is in the composite conflict scene, a fourth multi-objective optimization function taking the line safety performance index and the conflict point safety performance index as main optimization sub-objectives and taking the traffic efficiency index, the energy economy index and the riding comfort index as auxiliary optimization sub-objectives is adopted.
6. The vehicle scheduling method of claim 1, wherein the safety constraints include a conflict point minimum safety time interval constraint, a same lane minimum safety distance constraint, a vehicle speed physical feasibility domain constraint, an acceleration physical feasibility domain constraint, and a jerk physical feasibility domain constraint.
7. The vehicle scheduling method according to claim 2, characterized in that the equivalent control input amount is a difference between an actual acceleration and a deceleration caused by rolling friction.
8. A vehicle scheduling apparatus, characterized by comprising: the acquisition module is used for acquiring the running state information from the running of the vehicle to the running before the intersection without the traffic light, and acquiring the relative positions of other vehicles in the current intersection area and the predicted running path information; The system comprises a selection module, a self-vehicle detection module and a prediction module, wherein the selection module is used for determining a corresponding multi-objective optimization function based on a traffic interaction scene where the self-vehicle is located in each sampling period, wherein the traffic interaction scene is determined based on the driving state information of the self-vehicle and the relative positions of other vehicles and predicted driving path information; the construction module is used for constructing a vehicle longitudinal dynamics model based on the running state information of the vehicle, and constructing a model predictive control problem by combining the multi-objective optimization function and the safety constraint condition; And the control module is used for solving the model predictive control problem, obtaining a target control input sequence in a control time domain, and outputting the first control input quantity in the target control input sequence to the execution mechanism of the self-vehicle so as to realize real-time regulation and control of the self-vehicle.
9. A vehicle, characterized in that the vehicle comprises a processor and a memory, the memory storing a computer program, the processor being adapted to execute the computer program to implement the vehicle scheduling method of any one of claims 1-7.
10. A computer readable storage medium, characterized in that it stores a computer program which, when executed on a processor, implements the vehicle scheduling method according to any one of claims 1-7.

Description

Vehicle scheduling method and device, vehicle and storage medium Technical Field The present application relates to the field of vehicle control technologies, and in particular, to a vehicle scheduling method, a vehicle scheduling device, a vehicle, and a storage medium. Background The traffic intersection is used as a key node for most frequent vehicle intersection and diversion in a road network, and is a key link of urban traffic flow organization. Because the multidirectional traffic flows are converged at the height, traffic collision is easy to cause, and traffic jam is aggravated and delayed to increase, the multidirectional traffic flows become traffic bottlenecks in the whole traffic system, and particularly under the condition of no signal lamp or signal lamp damage. Meanwhile, the complex interaction behavior also makes the intersection become a high-incidence area of traffic accidents, and a serious challenge is formed to driving safety. Therefore, how to realize safe and orderly passing of vehicles at intersections without signal lamps is not only the key point for improving the running efficiency of roads, but also the important problem to be solved for propelling urban automatic driving and landing. Disclosure of Invention In view of this, embodiments of the present application provide a method, a vehicle scheduling apparatus, a vehicle, and a computer-readable storage medium. In a first aspect, an embodiment of the present application provides a method, including: acquiring running state information before a self-vehicle runs to an intersection without traffic lights, and acquiring relative positions of other vehicles in the current intersection area and predicted running path information; In each sampling period, determining a corresponding multi-objective optimization function based on a traffic interaction scene where a vehicle is located, wherein the traffic interaction scene is determined based on the running state information of the vehicle and the relative positions and predicted running path information of other vehicles; Constructing a longitudinal dynamics model of the vehicle based on the running state information of the vehicle, and constructing a model predictive control problem by combining the multi-objective optimization function and the safety constraint condition; Solving the model predictive control problem to obtain a target control input sequence in a control time domain, and outputting the first control input quantity in the target control input sequence to an execution mechanism of the own vehicle to realize real-time regulation and control of the own vehicle. In an alternative embodiment, the building a vehicle longitudinal dynamics model based on the running state information of the own vehicle includes: Based on a nonlinear continuous dynamic equation of longitudinal movement of the vehicle, carrying out local linearization processing at the initial state of the current sampling period to obtain a linearized state space expression; discretizing the state space expression to obtain a discrete time state space model; And combining the ground rolling friction resistance item into a control input variable to form an equivalent control input quantity, and assuming that a system state transition matrix and the control input matrix are kept unchanged in a prediction time domain so as to construct the longitudinal dynamics model of the vehicle, wherein displacement and speed are used as state variables, and the equivalent control input quantity is used as a control variable. In an alternative embodiment, constructing model predictive control problems based on the vehicle longitudinal dynamics model in combination with the multi-objective optimization function, and the safety constraints, includes: deriving a predicted state sequence of the own vehicle in a prediction time domain based on the longitudinal dynamics model; Converting the safety constraint into a set of inequality constraints with respect to the state prediction sequence and the control sequence; And constructing the model predictive control problem based on the multi-objective optimization function and the inequality constraint set. In an alternative embodiment, the traffic interaction scene includes a collision-free scene, a linear collision scene, a punctiform collision scene, and a composite collision scene: the collision-free scene is a risk scene in which potential collision does not exist between the own vehicle and other vehicles; the linear conflict scene is a linear conflict scene in which the self-vehicle and other vehicles on the same entrance lane have rear-end collision risks; the punctiform conflict scene is a punctiform conflict scene that the self-vehicle and other vehicles from different entrance lanes have paths crossing in the crossing; The composite conflict scene is a composite conflict scene containing both linear conflict risks and punctiform conflict risks. In an optional embodiment, th