Distributed robust consensus control for nonlinear leader-follower multi-agent systems based on adaptive observer-based sliding mode

This paper investigates an adaptive observer-based consensus control strategy for general nonlinear multi-agent systems (MASs). The distributed control of leader-follower MASs is studied in the presence of unknown parts in the dynamical equations of the followers which may be due to model simplification, parameter uncertainties and/or external disturbances. In order to decrease the conservatism of robust analysis, the upper bound of unknown terms are considered as unknown positive constants and are obtained through the adaptation laws. Additionally, it is assumed that all the agents' states are not directly measurable and nonlinear distributed observers are designed. Furthermore, the robust consensus is achieved via distributed adaptive observer-based sliding mode controllers. In this regard, a theorem is given and it is proved that the consensus error converges to zero. Finally, computer simulations are performed and the theoretical results are verified.