Car‐following strategy of intelligent connected vehicle using extended disturbance observer adjusted by reinforcement learning

Disturbance observer-based control method has achieved good results in the car-following scenario of intelligent and connected vehicle (ICV). However, gain conventional extended disturbance observer (EDO)-based is usually set manually rather than adjusted adaptively according to real time traffic conditions, thus declining performance. To solve this problem, a strategy ICV using EDO by reinforcement learning proposed. Different from method, proposed can be improve its estimation accuracy. Since “equivalent disturbance” compensated great extent, rejection ability will improved significantly. Both Lyapunov approach numerical simulations are carried out verify effectiveness method.