Two-Stage Deployment Strategy for Wireless Robotic Networks via a Class of Interaction Models

Suppose a disaster happens, and several groups of robots are dispatched from distant control stations. To enable rescue staff to make collective decisions, reliable and robust connections need to be established among stations. Motivated by this scenario, a two-stage robot deployment strategy is proposed for wireless robotic networks (WRNs). In the first stage, robots in distant groups are merged into one group that covers a desirable area. Since connectivity alone cannot guarantee a high communication quality, in the second stage, the flow between any two stations is further optimized in terms of expected number of transmissions per successfully delivered packet. In both stages, a distributed collision-free controller is proposed to regulate the interactive force among robots. The stability issues of WRNs, where the proposed controller together with a class of interaction models based on an acute angle test is implemented for robots, are analyzed under both fixed and switching topology. In order to efficiently switch the neighbor set for each robot, a new energy function is constructed taking finite-time consensus into account. To guarantee that energy agreement is achieved before the next topology change, a fixed-time consensus approach is further proposed and an upper bound of the settling time for the energy agreement is obtained. Numerical simulations are provided to demonstrate the effectiveness of the two-stage deployment strategy.