Multi-Agent Localization Using Geometric Constraints with Intermittent State Feedback

Mobile agents are often tasked with following paths in
environments where state feedback may not always be available, such as in GPS denied environments, and more broadly, anti-access and area-denial environments (A2/AD). Such applications have given rise to a class of Relay-Explorer problems, where an agent is tasked with switching between multiple operation modes, primarily depending on the availability of state feedback. Past works have established frameworks for developing dwell-time conditions and path planning techniques to guarantee stability, for systems where the objective requires operation in locations where feedback is unavailable. This paper extends the study of Relay-Explorer problems to multi-agent systems so cooperative localization can enhance the path tracking performance of each agent and extend the duration the ensemble of agents can spend operating without state feedback. To accomplish these goals, a new cooperative localization method is developed that uses the agents’ dynamics and the distance measurements between them. Hybrid system theory is used to model and analyze the proposed localization method and path planning strategy. In a comparative simulation study the cooperative behavior is shown to yield significant