Persistent surveillance for heterogeneous robots considering movement randomness and energy allocation.

This paper addresses a new persistent surveillance problem where heterogeneous robots measure task nodes, taking into account movement randomness and energy allocation. The heterogeneous robots, which consist of an unmanned ground vehicle (UGV) and unmanned aerial vehicles (UAVs), are deployed in real-time counteracting environmental inspection applications. To avoid hostile intruders from predicting their future routing information and to improve measurement efficiency, this paper seeks to: (1) address the privacy of persistent surveillance, and (2) make balanced programs about the energy cost. Specifically, this paper first proposes a framework in which UAVs perform stochastic movement based on Markov Chain and leverage the probabilistic measurement to fulfill the overall frequency of the task nodes. Then, persistent surveillance with movement randomness and energy allocation (PSREA) is formulated as an optimization problem, which is non-convex and becomes quite complex when regarding excessive task nodes. A clustering-based task network simplification algorithm and an iterative two-stage algorithm are proposed to cover an enormous number of task nodes and to deal with the non-convex problem, respectively. The numerical results demonstrate that the proposed algorithms can significantly improve the inspection performance compared to the benchmark results.