Intelligent $H_{\infty}$ Control for UAVs via Fuzzy Deep Reinforcement Learning

The problem of intelligent
$H_{\infty}$
control for unmanned aerial vehicles (UAVs) is investigated in this paper. The linear model of UAV can be obtained by the aid of Jacobian linearization. Considering the time delay and packet losses in the network, the robust
$H_{\infty}$
controller is proposed to overcome the undesirable response caused by time delay and packet losses. The deep reinforcement learning is introduced to improve the performance. Moreover, to improve the learning efficiency, we utilize a fuzzy reward system for the control process. The non-fragile control theory and Lyapunov functional method are combined to ensure the stability of closed-loop system. Simulation results in the end demonstrate the effectiveness and superiority of proposed method.