Distributed Adaptive NN Resilient Optimal Control for Heterogeneous Vehicular Platoon Systems Under DoS Attacks

Vehicular platoon systems are multiple intelligent vehicles travelling longitudinally and maintaining a desired inter-vehicle spacing. In this paper, a distributed adaptive neural network (NN) resilient optimal control problem is investigated for heterogeneous vehicular platoon systems (VPSs) subject to denial-of-service (DoS) attacks. Since the communication channels are suffered from DoS attacks, the leader's information cannot be continuously obtained by the heterogeneous VPSs, a distributed resilient filter is utilized to estimate unknown leader. Based on the designed distributed resilient filter and the differential graphical game strategy, a distributed adaptive NN resilient optimal control scheme is formulated through a sliding mode surface. The developed resilient optimal control scheme can ensure the following vehicles can asymptotically track the leader, and obtain the global Nash equilibrium solution of the differential graphical game strategy. Finally, the validity of the proposed resilient optimal control scheme is demonstrated by simulation.