A sliding mode control strategy for vehicle active steering optimized by Runge-Kutta algorithm

With the rapid advancement of vehicle intelligence and electrification, the Active Front Steering (AFS) system has emerged as a pivotal technology for enhancing vehicle handling stability and safety. However, the performance of Sliding Mode Control (SMC), which is widely applied in AFS, is heavily contingent upon controller parameters. Conventional manual tuning methods are empirical, laborious, and often fail to guarantee optimal robustness across diverse driving conditions. To address these limitations and bridge the gap between theoretical control design and engineering application, this study aims to develop a novel autonomous parameter tuning strategy for AFS systems. A co-design optimization framework is established by integrating a linear two-degree-of-freedom (2-DOF) vehicle dynamics model with an SMC controller. The Runge-Kutta Optimizer (RUN) is employed to precisely optimize the sliding surface and reaching law parameters by minimizing a stability-based objective function. The proposed strategy is validated through a MATLAB/Simulink and CarSim co-simulation platform under a double lane change scenario. Comparative results demonstrate that the proposed RUN-based approach significantly outperforms benchmark algorithms. Specifically, compared to Particle Swarm Optimization (PSO) and Grey Wolf Optimizer (GWO), the convergence speed is accelerated by 42.3% and 35.7%, while the solution accuracy is enhanced by 28.6% and 21.4%, respectively. These findings confirm that the proposed methodology substantially elevates the dynamic stability of the vehicle.