Asymptotic Leader-Following With Continuous Control Inputs in Networked Euler-Lagrangian Systems: The Case of Directed Topologies

In most practical dynamic leader-following scenarios, the control input or dynamical model of the neighboring agents is not known to the followers. In such scenarios, asymptotic leader-following requires employing discontinuous robust mechanisms to cope with unknown parts of the neighbors’ trajectories, leading to chattering in the followers’ inputs. Few studies have already addressed asymptotic leader-following with continuous control inputs; however, they require the communication topology to be undirected, which is a restrictive assumption in leader-following. In this article, asymptotic dynamic leader-following in Euler–Lagrangian multiagent systems is addressed. We assume that the leader’s control input is determined by the leader locally and is unknown to the followers. The contribution of this article is that whereas the unknown input and model of the neighboring agents are compensated by a robust mechanism in the followers’ control strategy, the continuity in the followers’ control inputs is guaranteed, and the interaction in the network is under directed topologies. Simulation results for a network of manipulator robots illustrate the performance of the proposed control strategy.