Cooperative Game-Based Optimal Shared Control of Unmanned Aerial Vehicle

Collaboration between human operators and unmanned aerial vehicles (UAV) is an emerging research area in UAV control, focused on enhancing UAV performance and alleviating the workload of human operators. This paper proposes a dynamic event-triggered cooperative control method for the optimal shared control of UAV. The proposed method formulates a cooperative nonzero sum game between human operators and UAV, and the optimal shared control of UAV is achieved by solving the Nash equilibrium of the game. The optimal shared control of UAV is approximated by the actor-critic algorithm, which learns the optimal control policy from the historical operation data of UAV. Then a continuous and efficient shared mechanism is established to allocate the relationship between optimal and human control input. To reduce the online computation burden of the optimal shared control in the UAV system, a dynamic event-triggered mechanism is proposed to update the shared control input only when the event-triggered condition is satisfied. The performance of the proposed method is evaluated by numerical simulations. The results show that the proposed method could achieve optimal shared control of UAV with human operators efficiently and continuously compared with conventional shared control methods.