Coordinating networked uninhabited air vehicles for persistent area denial

Abstract

Abstract-This paper explores the problem of cooperative control among multiple networked unmanned air vehicles (UAVs) for persistent area denial (PAD) mission. An adaptive Markov chain model is used to predict the locations of pop-up threats. The mixed information of predicted pop-up threats and actual pop-up targets is utilized to develop cooperative strategies for networked UAVs. The approach is illustrated by use of a simulation test bed for multiple networked UAVs and Monte Carlo simulation runs to evaluate our cooperative strategy. Both theoretical analysis and simulation results are presented to demonstrate the effectiveness of using predicted pop-up information in improving the overall PAD mission performance. I. INTRODUCTIONNINHABITED air vehicles have been identified as potential valuable resources for future military, communications and earth-science efforts. Cooperation among multiple UAVs is a key capability for utilizing the full potential of UAV systems. With respect to military applications, one of the potential missions for UAVs is persistent area denial. The aim of multiple UAVs in PAD operations is to provide persistent surveillance, tracking, and rapid engagement with high-volume strike, against threats (e.g., enemy integrated air defense system) at various ranges in the adversarial terrains. Threats in the battle space, in general, are of two types: known and popup. The locations of known threats are identified by the UAVs at the beginning of the battle, while the locations of pop-up targets are not always observable during the entire PAD operation. In other words, the pop-up targets may appear and disappear at frequent and random intervals of time. Obviously, uncertainty introduced by pop-up targets presents the significant theoretical and technical challenges This work was supported by the AFRL/VA and AFOSR Collaborative Center of Control Science Utilizing a mixed integer programming approach, a cooperative control strategy of multiple UAVs for PAD operation is developed in [2] as well. In this strategy, the trajectories of vehicles are optimized so that they can reach all targets in the shortest time. The predicted information, however, is not utilized in the computation of cooperative control strategies. In this paper, we modify this cooperative strategy in a manner that allows the UAVs to make use of both actual and predicted information about pop-up targets in coordinating their target assignments.Moreover, since recent advances in telecommunication have provided enabling technologies for achieving cooperative control of multiple UAVs via a communication network [3], we will consider networked UAVs in performing the PAD operations. II. SYSTEM MODEL A. Adaptive Markov-Chain Model of Pop-up TargetsPop-up targets are ground assets that appear at unpredictable locations and at random instants of time. In addition, pop-up targets may stay for a regular interval time and then disappear. Suppose that the location of the next pop-up target only depends on the location of present targets in the...