Trajectory Tracking and Yaw Stability Control for Electric Vehicles Based on LTV-MPC

This study focuses on Linear Time-varying Model-based Predictive Control (LTV-MPC) to support real-time trajectory tracking and yaw stability control for distributed drive electric vehicles (DDEV) under steering conditions. First, the nonlinear vehicle dynamics model was transformed into a linear model through local linearization. Then, a multi-objective optimization function was developed, considering multiple control objectives: trajectory tracking based on lateral positions and yaw angles; yaw stability based on yaw rates and side slip angles; power performance based on longitudinal speeds. In this function, the side slip angle of tires is defined as a soft constraint to avoid yawing force saturation. These optimizations involve front wheel steering angles, total longitudinal driving force, and additional yawing moments, contributing to effective motion control for the vehicles. Finally, the effectiveness and real-time performance of the proposed control strategy were verified using a simulation platform that combines CarSim/Simulink and Ni PXI system. Simulation results showed that the trajectory tracking errors from the LTV-MPC control method closely aligned with those from offline control, while the response times remained almost unchanged for key parameters such as yawing angles, longitudinal speeds, and front wheel steering angles. Furthermore, the amplitude of yaw angles achieved by the LTV-MPC control method was about 20% less than that based on offline control, which indicates that the LTV-MPC control method is more effective in ensuring yaw stability for these vehicles.