Research on MPC-based Control of Electric Active Anti-roll Bar

To enhance vehicle lateral stability and anti-roll performance, a model predictive controller (MPC) is designed for an electrically driven active anti-roll bar. First, a vehicle roll dynamics model and an active anti-roll bar model based on motor drive are established. Then, model predictive controllers for the front and rear suspension anti-roll bars are designed to calculate the anti-roll moments for the front and rear suspensions. Finally, co-simulation using Carsim and Simulink is conducted to simulate typical vehicle driving scenarios. The simulation results of passive suspension, PID and MPC control are compared to validate the improvement in vehicle stability achieved by MPC control, and the sensitive parameters of the controller affecting anti-roll performance are analyzed. The research results show that, compared to passive suspension and traditional PID control, the maximum roll angle with model predictive control is reduced by 80% and 40%, respectively. The MPC controller exhibits faster response speed and better anti-roll performance. Although the MPC control outputs 8% more torque than PID control, it provides a 40% improvement in performance, achieving an optimal balance between system performance and control input. Consequently, the energy consumption of MPC control is lower. The prediction horizon and performance weighting coefficients are identified as sensitive parameters of the controller.