Experimental Analysis of Multi-Robot Network Resilience Strategy Using Control Barrier Functions

摘要

Resilience is an important property for multiple robots to achieve coordination in the presence of communication and environmental interference. With the existence of Byzantine robots such as malfunctioning or malicious robots disrupting the coordination strategy, multi-robot network resilience depends on the designed control strategy in an unstructured environment with obstacles. In this paper multi-robot resilient control strategy based on control barrier functions (CBFs) is proposed, where network resilience, collision and obstacle avoidance constraints are considered at the same time. These conflicting objectives are trade-off by the control strategy. Physical experiments for two types of scenarios such as consistency about headings of multi-robots and consistency about the estimated state of target are carried out, which are designed to analyze several factors affecting multi-robot network resilience with the adopted strategy. The distributed implementation of resilience strategy and some helpful experiences for future works are discussed.