重庆理工大学学报2025，Vol.39Issue(9)：10-19,10.DOI:10.3969/j.issn.1674-8425(z).2025.05.002
基于GA-PSO优化的汽车轨迹跟踪和稳定性协同控制
Trajectory tracking and stability collaborative control for vehicle based on GA-PSO
摘要
Abstract
For intelligent vehicles and autonomous driving,maintaining precise trajectory tracking and vehicle stability is crucial for safety.However,under adverse operating conditions such as low road adhesion(e.g.wet or icy roads)or high-speed maneuvers,achieving high trajectory tracking accuracy and stability becomes challenging due to factors like tire slippage and dynamic uncertainties.To address these issues,this paper proposes a solution based on a hierarchical control strategy that coordinates trajectory tracking and stability control. The proposed control strategy employs a layered control architecture consisting of an upper-level trajectory tracking controller and a lower-level direct yaw moment controller.The upper-level controller is responsible for ensuring the vehicle follows the desired path accurately.This is realized using Model Predictive Control(MPC),which optimizes the steering and velocity commands over a prediction horizon while considering vehicle dynamics and constraints.The lower-level controller maintains vehicle stability by regulating the yaw motion by employing Sliding Mode Control(SMC)to generate the required additional yaw moment and then allocate wheel torques on a distributed-drive vehicle according to optimal tyre-adhesion utilisation.Thus,it achieves torque distribution that meets the yaw-moment demand. To ensure robust performance in various driving conditions,key controller parameters for both the MPC and SMC layers must be appropriately tuned.To achieve this,a hybrid Genetic Algorithm-Particle Swarm Optimization(GA-PSO)algorithm is employed to optimize the controller parameters over a range of vehicle speeds and road adhesion coefficients.GA-PSO combines the global search capability of genetic algorithms with the efficient convergence of particle swarm optimization,making it well-suited for identifying near-optimal solutions in a complex parameter space.By conducting GA-PSO optimization offline over a broad range of speeds and adhesion levels and storing the results in an online interpolation table,the controller automatically adjusts the prediction horizon,control horizon,state-weight coefficients,and sliding-surface parameters according to real-time speed and adhesion estimates,thereby achieving adaptive parameter tuning. To verify the effectiveness of the proposed control strategy,a co-simulation testing environment is constructed using CarSim and Simulink.CarSim provides a high-fidelity vehicle dynamics model(capturing realistic tire forces,suspension dynamics,and other nonlinear behaviors),while the control algorithms are implemented in Matlab/Simulink.Simulation results demonstrate the GA-PSO-optimized hierarchical control strategy markedly improves path-tracking accuracy and yaw stability compared with pure MPC control and fixed-parameter hierarchical control.In the co-simulation tests,the proposed control reduces the average lateral tracking error by 89.9%,46.4%,and 43.3%in three representative driving scenarios compared to a pure MPC control scheme,indicating a major enhancement in trajectory-tracking precision.Meanwhile,the vehicle's stability is markedly improved.The controller helps prevent excessive lateral slippage,significantly lowering peak lateral acceleration and vehicle sideslip angle during aggressive maneuvers or low-adhesion conditions.These improvements highlight the GA-PSO tuned MPC/SMC coordinated controller effectively keeps the vehicle on the intended trajectory while minimizing instability under challenging driving conditions.Overall,the proposed solution improves the safety and performance of intelligent vehicles in more complicated environments.关键词
智能车辆/轨迹跟踪/稳定性控制/模型预测控制/滑模控制/遗传粒子群算法Key words
intelligent vehicles/trajectory tracking/stability control/model predictive control/sliding mode control/GA-PSO algorithm分类
交通工程引用本文复制引用
田韶鹏,吴思沛,王龙..基于GA-PSO优化的汽车轨迹跟踪和稳定性协同控制[J].重庆理工大学学报,2025,39(9):10-19,10.基金项目
广西科技重大专项(桂科AA22068063) （桂科AA22068063）
