A Cooperative Recovery Framework for Resilient Multi-Robot Swarm Operations Under Loss of Localization in Unknown Environments

Localization is one of the most important tasks for mobile robot operations. Without such capability, a robot may wander toward unsafe states and never complete a desired task. Such capability is even more important in multi-robot system (MRS) operations in which their motion is coordinated based on consensus schemes that leverage information from surrounding neighbors. Thus, in the event of compromised or malfunctioning on-board positioning sensing (e.g., due to cyber attacks or faults) on individual robots, the entire robotic system may be hijacked toward undesired states. In this work, we target this problem by proposing a decentralized framework where: i) robots with loss of localization capabilities detect the anomalous behavior then generate a notification signal within information exchanges to alert neighboring robots, and ii) neighboring robots leverage their mobility to aid in recovery allowing compromised robots to re-localize. Our framework is validated in simulations and lab experiments on proximity-based formations of homogeneous unmanned multi-robot swarms.