Robust Asymptotic and Finite-time Tracking for Second-order Nonlinear Multi-agent Autonomous Systems

This paper investigates consensus based distributed robust asymptotic and finite-time tracking control strategy for second-order multi-agent autonomous systems. The protocol design uses states of the neighboring agents with directed communication topology in the presence of uncertainty associated with the autonomous agents. Robust adaptive learning algorithm uses with the protocol design for each follower agent to learn and adapt bounded uncertainty associated with nonlinear dynamics of the follower agents. Adaptive learning protocol also integrates with the follower agents protocol to learn and adapt bounded input of the leader. Lyapunov method with Graph, classical sliding mode, and terminal sliding mode theory use to guarantee that the proposed distributed control design can reach an agreement and follow the states of the leader in both finite-time and asymptotic sense. Analysis shows that consensus based protocol can force the states of the followers sliding surface to track the states of the leader sliding surface in finite-time and remain there. The proposed distributed consensus protocol does not require the exact bound of the uncertainty associated with the follower agents. Also, the proposed protocol does not require the exact bound of the leader input as opposed to other distributed cooperative control designs. Evaluation results with comparison are presented to demonstrate the validity of the theoretical argument for the real-time applications.