Tube-MPC and Fuzzy Logic Driven Control Strategy for Adaptive ARS-DYC Coordination in Intelligent Vehicles

Addressing the challenge of enhancing intelligent chassis performance under extreme conditions, as framed by the CVCI2025 benchmark, this paper develops a coordinated controller integrating Active Rear Steering (ARS) and Direct Yaw Moment Control (DYC) via four in-wheel motors. Initially, key vehicle parameters are identified using a particle swarm optimization (PSO) algorithm based on the Magic Formula tire model. To handle inherent parameter uncertainties, a Linear Parameter-Varying (LPV) vehicle dynamics model is established. Leveraging this identified LPV model, a Tube-based Model Predictive Controller (TMPC) is designed to compute the optimal ARS and DYC inputs. Crucially, a fuzzy logic method dynamically adjusts the weight matrices of the TMPC, optimizing the trade-off between vehicle stability and handling maneuverability. Comprehensive simulation studies demonstrate the proposed scheme's effectiveness, confirming significant improvements in both stability and overall handling performance under demanding driving scenarios.