CN-121979224-A - Two-dimensional formation composite sliding mode control method under explicit convergence time limit

CN121979224ACN 121979224 ACN121979224 ACN 121979224ACN-121979224-A

Abstract

The invention provides a two-dimensional formation composite sliding mode control method under explicit convergence time limit, and relates to the technical field of automatic driving cooperative control. The method comprises the steps of firstly obtaining real-time motion state information of a pilot vehicle and N following vehicles, establishing a third-order nonlinear dynamics model of each following vehicle, calculating tracking errors of the following vehicles based on virtual vehicle strategies, dynamically processing preset performance constraints, constructing a composite limited time sliding mode surface of a layered structure, determining total execution time limit required by all monitoring signals in a two-dimensional vehicle formation control system to converge from an initial deviation state to a predefined steady-state area, and finally estimating an external disturbance boundary on line based on the composite limited time sliding mode surface and combining an adaptive control algorithm to generate a control instruction to force the monitoring signals to reach a steady state in an explicit time upper boundary. The method provides an intuitive performance estimation means for practical engineering application, and improves the stability, the rapidity and the accuracy of two-dimensional formation control.

Inventors

CHEN JIAXIN
LI YUMENG
GAO ZHENYU

Assignees

东北大学秦皇岛分校

Dates

Publication Date: 20260505
Application Date: 20260326

Claims (8)

1. The two-dimensional formation composite sliding mode control method under the explicit convergence time limit is applied to vehicle formation in a two-dimensional plane and is characterized by comprising the following steps of: step S1, formation full-state information sensing and modeling, namely acquiring real-time motion state information of a pilot vehicle and N following vehicles, and establishing a third-order nonlinear dynamics model of each following vehicle; Step S2, calculating tracking errors of the following vehicles based on the virtual vehicle strategies, setting corresponding virtual vehicle reference points for each following vehicle, and calculating longitudinal distance tracking errors of the following vehicles according to the relative geometric relationship between the following vehicles and the corresponding virtual vehicle reference points Azimuth tracking error ; Step S3, dynamically processing preset performance constraint, constructing a finite time performance function related to the longitudinal distance tracking error amplitude, actively adjusting the performance boundary according to the current longitudinal distance tracking error, and converting the limited longitudinal distance tracking error into an unlimited transformation error through error transformation ; Step S4, according to the unrestricted transformation error Azimuth tracking error Constructing a composite finite time slip form surface of a layered structure On-line quantitative pre-estimation is carried out on the execution time limit of the two-dimensional vehicle formation control system signal converging to a predefined area through a preset explicit time mapping function based on the initial state of the two-dimensional vehicle formation control system and the design parameters of the controller, and the total execution time limit required by the convergence of all monitoring signals in the two-dimensional vehicle formation control system from the initial deviation state to the predefined steady-state area is determined ; S5, outputting and executing a cooperative control law, and generating a longitudinal control instruction based on a composite limited time sliding mode surface and combining an adaptive control algorithm to estimate an external disturbance boundary on line Steering control command Forcing the monitor signal to be at the explicit time upper bound Steady state is reached.
2. The two-dimensional formation composite sliding mode control method under explicit convergence time limit according to claim 1, wherein the three-order nonlinear dynamics model of the following vehicle takes into consideration an engine time constant, an air resistance coefficient, a road resistance coefficient and external environmental disturbance, and is expressed as: ; Wherein, the , , Following vehicles at times t respectively A longitudinal position, a lateral position and a heading angle, For vehicles Is used for the quality of the (a), 、 Following vehicles at times t respectively Is used for the speed and acceleration of the vehicle, 、 Following vehicles at times t respectively Is provided with a plurality of angular velocities, For following vehicles Is set to be a constant of the engine time, And Following vehicles at times t respectively Is provided with longitudinal and steering control commands, As a result of the known non-linear term, And Following the vehicle for time t Is a concentrated disturbance outside the longitudinal and steering directions of the vehicle.
3. The method for controlling a two-dimensional formation composite sliding mode under explicit convergence time limit according to claim 2, wherein the coordinates of the virtual vehicle reference point are determined according to a pilot vehicle track and a preset safety interval And the following vehicle geometric dimension parameters are calculated and generated in real time, and the following formula is shown: ; ; wherein, the method comprises the following steps of , ) For following vehicles Corresponding virtual vehicle reference point coordinates [ ] , ) For the position coordinates of the pilot vehicle, For following vehicles With the front vehicle The desired safe distance between-1, For following vehicles Is the distance of the centroid of (c) from its front end, For the front vehicle -1 From the rear end thereof; According to following vehicles Calculating the following vehicle according to the relative geometrical relation with the corresponding virtual vehicle reference point Tracking error of longitudinal pitch of (2) Azimuth tracking error The following formula is shown: ; ; ; ; wherein, the method comprises the following steps of , ) For following vehicles Is defined by the centroid coordinates of (a), For following vehicles The actual distance between its corresponding virtual vehicle, For following vehicles Azimuth angles between their corresponding virtual vehicles, For following vehicles Is used for the actual course angle of the car.
4. The method for two-dimensional formation composite sliding mode control under explicit convergence time limit according to claim 3, wherein said step S3 converts limited tracking error into non-limited transformation error by error transformation The transformation logic of (a) is: when the initial longitudinal pitch tracking error is positive, the transformation logic is as follows: ; when the initial longitudinal pitch tracking error is negative, the transformation logic is as follows: ; Wherein, the , , And Are all based on longitudinal spacing tracking error A finite time performance function of amplitude dynamic adjustment; tracking error for longitudinal spacing Constraint conditions are introduced, and the following formula is shown: ; Wherein, the And Is a lower and upper bound of a predefined longitudinal pitch tracking error boundary; During the duration of Inside, longitudinal pitch tracking error Stable within a given bounded region, as shown by the following formula: ; Wherein, the And Is any given normal number, meets ; Longitudinal pitch tracking error The following performance constraints are met: ; Wherein, the For following vehicles Is used for tracking errors of the initial longitudinal spacing of the lens.
5. The method for two-dimensional formation composite sliding mode control under explicit convergence time limit of claim 4, wherein the composite finite time sliding mode surface of the hierarchical structure The following formula is shown: ; ; ; Wherein, the The system is a sliding mode variable and is used for representing the tracking state of a two-dimensional vehicle formation control system; Is that Is used for the purpose of determining the derivative of (c), , Based on transform errors The surface of the basic sliding mould is constructed, Is based on azimuth tracking error A constructed basic sliding mode surface, and ; Is a positive gain coefficient and ; As a nonlinear function containing fractional order power terms, In order to switch the threshold coefficient of the value, Is a coefficient of a linear term which is a coefficient of a linear term, Is a quadratic coefficient.
6. The explicit convergence time-limited two-dimensional formation composite slip-form control method as set forth in claim 5, wherein a total execution time limit required for all monitored variables within the two-dimensional vehicle formation control system to converge from an initial bias state to a predefined steady-state region Is explicitly calculated as: ; Wherein, the , , The time for the monitoring variable to converge from the initial energy state to the sliding mode surface, the sliding time of the sliding mode variable signal along the sliding mode surface and the time for the disturbance estimation error to enter a steady-state area are respectively; wherein the monitoring variable is a composite deviation signal comprising a composite finite-time sliding mode surface And disturbance estimation errors.
7. The method for two-dimensional formation composite sliding mode control under explicit convergence time limit according to claim 6, wherein the adaptive control algorithm performs real-time online estimation on the boundary of unknown external disturbance by constructing an adaptive law, and compensates the estimated value into a control command, and the generated longitudinal control command The method comprises the following steps: ; steering control command The method comprises the following steps: ; Wherein, the For the error transform coefficient(s), As the heading angle-acceleration correlation coefficient, 、 The longitudinal channel is respectively combined with the limited time sliding die surface and the azimuth channel is respectively combined with the limited time sliding die surface, 、、 For a predetermined gain constant to be set, Is a fractional order index, satisfies 0< <1, For the preset stability adjustment coefficient, the control device, 、 Longitudinal disturbance boundary estimation values and azimuth disturbance boundary estimation values are respectively adopted, 、 Are compensation functions.
8. The method for controlling a two-dimensional formation composite sliding mode under explicit convergence time limit according to claim 7, wherein before constructing the composite limited-time sliding mode, the method further ensures chord stability of the formation based on a coupling variable technology, and defines a coupling error between adjacent vehicles: ; Wherein, the For vehicles Coupling errors with the neighboring vehicles, For the preset stability adjustment coefficient, the stability adjustment coefficient is selected to meet 0 The stability adjustment coefficient of (2) ensures that the disturbance is suppressed during the backward propagation of the fleet; and constructing a basic sliding mode surface based on the coupling error, constructing a composite limited time sliding mode surface by the basic sliding mode surface, and finally generating a longitudinal control instruction.

Description

Two-dimensional formation composite sliding mode control method under explicit convergence time limit Technical Field The invention relates to the technical field of automatic driving cooperative control, in particular to a two-dimensional formation composite sliding mode control method under explicit convergence time limit. Background With the development of automatic driving technology, vehicle formation control has become an important research direction in an automatic driving system, and the current formation control research has been expanded from a one-dimensional plane to a two-dimensional plane which is more in line with actual traffic scenes, such as multi-lane junction, lane change and the like. However, the existing two-dimensional formation control technology still has the defects that the heading response is redundant, the existing method requires vehicles to track adjacent forward traveling, unnecessary heading angle adjustment is generated in the course of changing lanes or converging, energy consumption is increased, the performance constraint is single, the traditional preset performance control function is independent of tracking errors, balance between system singularities and acceleration convergence speed is difficult to achieve, the convergence time is invisible, and even under a limited time control framework, the specific upper limit of the convergence time of the system is difficult to be expressed explicitly through initial states and control parameters, and an intuitive performance prediction means is lacked. Therefore, it is necessary to design a two-dimensional formation composite sliding mode control method and system under explicit convergence time limit. Disclosure of Invention The invention aims to solve the technical problems of the prior art, and provides a two-dimensional formation composite sliding mode control method under explicit convergence time limit, which is used for solving the problems of redundant course response, single performance constraint and invisible convergence time of the two-dimensional formation control technology. In order to solve the technical problems, the technical scheme adopted by the invention is that the two-dimensional formation composite sliding mode control method under the explicit convergence time limit is applied to vehicle formation in a two-dimensional plane and comprises the following steps: step S1, formation full-state information sensing and modeling, namely acquiring real-time motion state information of a pilot vehicle and N following vehicles, and establishing a third-order nonlinear dynamics model of each following vehicle; Step S2, calculating tracking errors of the following vehicles based on the virtual vehicle strategies, setting corresponding virtual vehicle reference points for each following vehicle, and calculating longitudinal distance tracking errors of the following vehicles according to the relative geometric relationship between the following vehicles and the corresponding virtual vehicle reference points Azimuth tracking error; Step S3, dynamically processing preset performance constraint, constructing a finite time performance function related to the longitudinal distance tracking error amplitude, actively adjusting the performance boundary according to the current longitudinal distance tracking error, and converting the limited longitudinal distance tracking error into an unlimited transformation error through error transformation; Step S4, according to the unrestricted transformation errorAzimuth tracking errorConstructing a composite finite time slip form surface of a layered structureOn-line quantitative pre-estimation is carried out on the execution time limit of the two-dimensional vehicle formation control system signal converging to a predefined area through a preset explicit time mapping function based on the initial state of the two-dimensional vehicle formation control system and the design parameters of the controller, and the total execution time limit required by the convergence of all monitoring signals in the two-dimensional vehicle formation control system from the initial deviation state to the predefined steady-state area is determined; S5, outputting and executing a cooperative control law, and generating a longitudinal control instruction based on a composite limited time sliding mode surface and combining an adaptive control algorithm to estimate an external disturbance boundary on lineSteering control commandForcing the monitor signal to be at the explicit time upper boundSteady state is reached. Further, the third-order nonlinear dynamics model of the following vehicle considers an engine time constant, an air resistance coefficient, a road resistance coefficient and external environment disturbance, and is expressed as: ; Wherein, the ,,Following vehicles at times t respectivelyA longitudinal position, a lateral position and a heading angle,For vehiclesIs used for the quality of the (a