Observer-Based Robust Explicit Model Predictive Control for Path Following of Autonomous Electric Vehicles with Communication Delay

• Proposes a projected interval unscented Kalman filter (PIUKF) that incorporates state constraints to accurately estimate vehicle sideslip angle and yaw rate, enhancing robustness. • Develops an explicit model predictive control (EMPC) strategy that significantly reduces online computational burden compared to traditional MPC, enabling real-time implementation. • Integrates communication delay compensation into the observer-based control framework, improving path following performance under realistic data transmission conditions. • Validates the proposed strategy through numerical simulations and hardware-in-the-loop tests, demonstrating its effectiveness and practicability for autonomous electric vehicles.