Autonomous Vehicle Motion Control and Energy Optimization Based on Q-Learning for a 4-Wheel Independently Driven Electric Vehicle

Motion control and energy-saving optimization are the research hotspots in the field of autonomous vehicles. This study takes four-wheel independent drive (4WID) electric vehicles (EVs) in CDC 2023 Challenge. Aiming to address the issues of desired speed tracking, vehicle body motion control, and energy consumption minimization posed by the challenge, the vehicle driving resistance was analyzed, and a vehicle longitudinal dynamics model was established. The extended state observer (ESO) is utilized to estimate the model error, thereby making the dynamic model approximate the real system. A controller combining a linear quadratic regulator (LQR) and Q-learning is designed. The total torque of the vehicle is obtained by LQR method, and then the torque is distributed to the four wheels by Q-learning, so as to realize the tracking of speed and course, and ensure the smooth operation and energy-saving control of the autonomous vehicles. Simulation results indicate that the proposed control strategy can achieve precise speed tracking control and outperforms both the PID and fuzzy rule controllers in reducing vehicle body motion and energy consumption.