A Longitudinal and Lateral Coordinated Control Method of Autonomous Vehicles Considering Time-varying Delay

In order to improve the control accuracy and stability of autonomous vehicles under large delay conditions, the paper proposes a longitudinal and lateral coordinated control method considering fixed sensor delay, actuator lag, and timevarying CAN communication delay. An adaptive all-pass filter time-delay estimator (AAPF-TDE) based on finite impulse response (FIR) and its improved strategies are designed to realize accurate online estimation of CAN communication delay. According to the estimated delay, a longitudinal and lateral coupling time-delay dynamics model with prediction revision and delay augmentation is constructed. A model predictive controller (MPC) combined with Lyapunov asymptotic stability constraints is then designed. CarSim/Simulink co-simulation and vehicle experiment results show that the proposed controller can improve the vehicle stability effectively while ensuring the longitudinal and lateral control accuracy under large delay conditions compared with the controller without considering delay or only considering fixed delay.