In this paper, a cooperative navigation algorithm for team of mobile agents is presented. The proposed algorithm is a loosely coupled decentralized algorithm. The setting we consider consists of a set of mobile agents with sensing, processing and communication capabilities. In this setting each agent maintains a local filter to propagate and update its local pose estimates using selfmotion measurements and occasional external absolute measurement signals. Whenever, an agent takes a relative measurement with respect to a teammate, our proposed algorithm enables these two agents to jointly process this relative measurements to update their local estimates in a consistent manner in the absence of exact knowledge about the correlation between their local estimates. Simulations demonstrate the benefits of our proposed algorithm.
Supplementary notes can be added here, including code and math. –>