MPC-based path tracking strategy for 4WID&4WIS vehicles using B-spline approximation and state-dependent reference

Abstract

Four-wheel independent drive and steering (4WID&4WIS) vehicles offer great potential for improving path tracking accuracy and vehicle stability. To fully leverage these capabilities, model predictive control (MPC) is widely adopted as an advanced path tracking strategy. However, conventional MPC approaches often struggle to simultaneously achieve high tracking accuracy and computational efficiency due to the trade-off between control horizon length and real-time feasibility. Furthermore, time-varying reference outputs can degrade tracking performance and even threaten vehicle stability, particularly when the vehicle deviates from predetermined temporal constraints. To address these challenges, this paper proposes an enhanced MPC strategy based on B-spline approximation and state-dependent reference (BS-MPC). Specifically, the control input sequence is approximated by a quasi-uniform B-spline curve, which significantly reduces the number of optimization variables and improves computational efficiency. Simultaneously, the reference output is adaptively generated as a function of the predicted vehicle state, which naturally couples longitudinal, lateral, and yaw motions and better aligns with the fundamental objectives of tracking control. Comprehensive Monte Carlo simulations validate the robustness and practical stability of the proposed BS-MPC controller under realistic uncertainty conditions. Finally, hardware-in-the-loop simulations and full-scale vehicle experiments demonstrate that, compared to conventional MPC, the proposed method reduces the maximum tracking errors by 86.2% under normal conditions and by 96.8% under extreme conditions.