Distributed, Cooperative Control for a Team of Autonomous, Tactical UAVs

Cooperative, autonomous control is required to realize the full potential of autonomous assets in a tactical operational setting. For a tactical battlespace, there are many levels of applications requiring cooperative control. This proposal addresses themission control level where the control task is to coordinate multiple heterogeneous distributed assets to achieve the desired global effect through the application of local force. When viewed as a centralized control problem under uncertainty, StochasticDynamic Programming (SDP) provides the optimal control strategy for this stochastic problem; however, this solution approach is not computationally feasible for most problems, and requires excessive centralization of information and computation. In thisproposal, we present innovative distributed control strategies that extend previous ALPHATECH Approximate Dynamic Programming (ADP) successes to cooperative control of tactical UAVs. The PHASE I work will extend these ADP strategies to the domain ofcooperative mission control of UAVs, and evaluate alternative distributed approaches for control coordination. The most successful approaches will be developed in detail in the PHASE II work.The proposed technology would enable autonomous heterogeneous distributed UAVs to cooperatively search, detect, identify, engage, and destroy enemy targets in a highly dynamic, uncertain environment, with limited human input. This concept of operationswould tailor the application of force, and thus achieve massed affects without relying on massed forces and sequential operations; as a result, a significant improvement in operational efficiency can be realized. In the commercial sector, this technologyis directly transferable to many applications involving autonomous systems, such as deep-sea exploration and salvage, hazardous material handling, flexible manufacturing systems and space-based assembly and repair.