Resilient Adaptive Multi-Robot Formation Tracking for Achieving Guaranteed Dynamic Obstacle Avoidance Within Limited Inter-Agent Communication Ranges

This paper presents a resilient adaptive formation control design strategy to achieve guaranteed dynamic obstacle avoidance within limited inter-agent communication ranges in nonholonomic multi-robot systems. Unlike existing literature, the main contribution lies in the development of a Lyapunov-based formation tracker utilizing guidance signals to avoid dynamic obstacles within limited inter-agent communication ranges in the presence of unknown deception attacks and unknown velocities of dynamic obstacles. Error variables, constructed using the guidance signals, are employed to design a resilient adaptive formation tracker within a neural network-based command-filtered backstepping framework. Through Lyapunov stability analysis, the guidance signals are designed to ensure that connected robots remain within limited inter-agent communication ranges during obstacle avoidance by adjusting formation errors that arise during obstacle avoidance. Additionally, adaptive laws are devised to compensate for unknown obstacle velocities and deception attacks affecting the leader’s velocity. The analysis guarantees that all error signals within closed-loop systems remain bounded and can be arbitrarily reduced in the Lyapunov sense. Simulation results confirm the effectiveness of the proposed theoretical framework.