Novel Extrapolating Methods Using Exponential Function in Formation Control of Swarm Robots

This paper addresses a position estimation problem for communication packet loss of a swarm of autonomous mobile robots. To complete assigned geometric patterns, these robots move while communicating each other. As is often the case in wireless communication, the robots cannot receive other robots' position data reliably. For the end, the robots need to overcome the packet-loss problem and the missed position data in near real-time. As the use of several generalized filer approaches (e.g., extended Kalman filter and particle filter) is considered, it would be difficult to be applied to low-cost robot swarms with limited computational capabilities and resources. Formally, as our solution to this problem, positional extrapolation schemes with reduced computational cost yet high estimation accuracy. By using the proposed scheme, our scheme enables the robot swarms to improve the accurate coverage performance despite less computation burdens. Other features include decentralization, convergence, and stabilization, which are proved mathematically and verified through extensive simulations. Our analysis and simulation results show that the scheme is a simple and efficient approach to the deployment of mobile robots.