Safety-critical target enclosing control for multi-robot systems: A robust reconfigurable strategy

This paper addresses the target enclosing control for multi-robot systems subject to actuator faults and external disturbances. First, a target enclosing framework is constructed based on bearing rigidity theory. Its formation shape is uniquely determined by inter-robot bearing measurements. Next, a fixed-time disturbance observer is designed to provide the estimation of lumped disturbances. To accommodate actuator physical constraints, a hierarchical fault-tolerant strategy is proposed. Driven by the disturbance estimates, the strategy adaptively triggers countermeasures including active compensation, leader reassignment, or robot detachment. Further, a quadratic programming-based controller is developed to guarantee system safety. It incorporates control barrier functions to handle obstacle avoidance and inter-robot collision avoidance under bounded estimation errors. Finally, numerical simulations and physical experiments demonstrate the effectiveness of the proposed robust reconfigurable algorithm.