On Nonlinear Model Predictive Vehicle Dynamics Control for Multi-Actuated Car-Semitrailer Configurations

Although the literature includes examples of model predictive controllers for trailer sway mitigation through direct yaw moment control in car-semitrailer configurations, the available studies do not consider the integration of direct yaw moment control and other actuations within the towing vehicle, e.g., anti-roll moment distribution control through active suspension (AS) systems. This study focuses on the integration of torque-vectoring (TV) and AS control on the towing car, to provide yaw rate tracking, sideslip angle limitation, and trailer sway reduction. Given the nonlinearity of the effect of the anti-roll moment distribution on the articulated vehicle dynamics, nonlinear model predictive control (NMPC) is selected as an effective technique for the specific problem. The paper presents and analyzes several alternative real-time implementable NMPC algorithms, using either the predicted trailer hitch dynamics or the estimated hitch joint forces for the control of an electric all-wheel-drive car towing a semitrailer. The simulation results, obtained through a high-fidelity simulation environment, including an optimization based tuning routine for a fair comparison of the controllers, show that: i) AS actuated in isolation brings higher semitrailer stability than a baseline TV controller focusing only on the towing car; ii) the best performance is provided by the integrated control of AS and TV, with respect to (w.r.t.) the other considered configurations; iii) the NMPCs based on car prediction models considering the current hitch joint forces can be a valuable alternative to the NMPCs embedding the hitch dynamics; and iv) all NMPC approaches are rather robust w.r.t. significant variations of semitrailer parameters.