Model Predictive Contouring Control for Vehicle Obstacle Avoidance at the Limit of Handling Using Torque Vectoring

• DOI: 10.48550/arxiv.2405.10847
• Publication Date: 2024-05-17
• Authors: Alberto Bertipaglia; Davide Tavernini; Umberto Montanaro; Mohsen Alirezaei; Riender Happee; Aldo Sorniotti; Barys Shyrokau
• Keywords: Model Predictive Control; Obstacle avoidance

This paper presents an original approach to vehicle obstacle avoidance. It involves the development of a nonlinear Model Predictive Contouring Control, which uses torque vectoring stabilise and drive in evasive manoeuvres at limit handling. The proposed algorithm combines motion planning, path tracking stability objectives, prioritising collision avoidance emergencies. controller's prediction model is double-track based on extended Fiala tyre capture coupled longitudinal lateral dynamics. controller computes optimal steering angle forces per each four wheels minimise error safe situations maximise vehicle-to-obstacle distance Thanks optimisation forces, can produce extra yaw moment, increasing vehicle's agility avoid obstacles while keeping stable. are constrained friction circle not exceed tyres capabilities. In high-fidelity simulation environment, we demonstrate benefits vectoring, showing that our capable successfully avoiding stable driving double-lane change manoeuvre, comparison baselines lacking or prioritisation.