Consensus Control for Multi‐Agent Systems With Input and Position Constraints

ABSTRACT This paper aims to achieve consensus for multi‐agent systems subject to convex position constraints and nonconvex input constraints. The main challenge arises from the coupling between position and input constraints, where the control input needs to be designed to preserve constraint satisfaction and guarantee consensus convergence. There is no common global benchmark to linearize the position constraints, and the simultaneous presence of position and input constraints may introduce new nonlinearities that cannot be eliminated. To solve this issue, we design an algorithm to make the input track the direction of the position variation by constructing a position estimation while ensuring that the positions and inputs remain inside the given constraint sets. The results show that consensus under position and input constraints can be effectively achieved even with switching topologies and communication delays. Finally, simulations are conducted to verify the effectiveness of the proposed approach.