Lateral and longitudinal path tracking control for intelligent electric vehicles based on tire cornering characteristics

Abstract

During vehicle operation, tire cornering characteristics vary dynamically with driving conditions and road conditions, making it difficult for fixed tire cornering stiffness (TCS) parameters to accurately represent the actual lateral dynamic characteristics of the vehicle. To improve the path-tracking accuracy of intelligent electric vehicles, this paper proposes a longitudinal-lateral coordinated path-tracking control method considering tire cornering characteristics. First, a vehicle dynamics model is established, and the concept of equivalent tire cornering stiffness (ETCS) is introduced. The forgetting factor recursive least squares (FFRLS) method is then employed to estimate and update the front and rear equivalent tire cornering stiffness online. Second, in lateral control, model predictive control (MPC) is adopted as the core controller, and the velocity pausing particle swarm optimization (VPPSO) algorithm is used to optimize the weighting matrix. In longitudinal control, a double-layer PID controller is employed to track the target speed. Finally, simulations are carried out under the double-lane-change maneuver. The results show that, compared with the fixed MPC-fixed TCS method, the proposed method reduces the lateral error by 62.4%; compared with the fixed MPC-FFRLS method, the lateral error is reduced by 41.9%. These results verify the effectiveness of the proposed method.