Cooperative Control of UAVs in Dynamic Pursuit Subject to Communication Delays

Abstract

An exponentially stable observer for MIMO nonlinear systems is designed and implemented in a simulation to evaluate the performance of a cooperative control law subject to known time delays. The cooperation among the vehicles is enforced through a potential fleld approach. Communication delays among cooperating vehicles is modeled as time-delays and a cascade of nonlinear observers exponentially estimate the positions and velocities of the neighboring vehicles. The performance of the control laws together with the observation process is evaluated on a planar vertical take ofi and landing aircraft. f late, there has been substantial amount of research being conducted in the area of cooperative control of multiple vehicles in achieving a common objective. The application areas have been several ranging from robot teams, micro-robot swarming, unmanned ground vehicles, unmanned aerial vehicles as well as micro-satellite clusters (Refs. [1{5]). These applications have further received a boost following signiflcant developments in control techniques for single vehicles, phenomenal improvements in computation and communication capabilities and the advent of miniaturization technologies. Continuing research has generated immense interest in vehicle systems that can interact autonomously with the operating scenario and other vehicles to cooperatively perform, tasks beyond the ability of individual vehicles in the face of highly unstructured and severe environments. It is envisioned that unmanned air vehicles (UAV) of the future will be more lethal and more autonomous than their remotely piloted reconnaissance platforms and several research issues have been identifled and are being actively pursued, that include dynamic coupling based on task assignments, mission re-planning, path planning and cooperative control among other open issues (See Refs. [2,6{9]). Of these issues pertinent to the discussion in this paper, UAV cooperative control with information ∞ow constraints was studied in Ref. [10] and Ref. [11]. In Ref. [10], three difierent situations were investigated that represent communication of information between vehicles as a sequence of impulses, a bandwidth-limited signal, and a range-limited signal. Further, the communication of information between vehicles was considered as another control input so that the UAV cooperative control problem with information ∞ow constraints could be formulated to be a series of generalized optimal control problems. Ref. [11] considers a broader range of vehicle interconnection possibilities for formation control to understand the efiect of topology of the information ∞ow on the stability and performance of the system as it performs a coordinated task. The authors develop information exchange strategies which improve formation stability and performance that are su‐ciently robust to changes in the communication topology by merging ideas from graph theory and systems theory. We note that formation control of cooperating vehicles depends heavily on information from the participating vehicles that may be subject to uncertainty and transmission delay. In an interconnected dynamical system the behavior of the participating systems depends not only on the individual vehicle dynamics, but on the nature of the interconnections, and hence it is essential to study the stability and performance of such systems 12 and the efiect of in∞uences that alter the nature of these interconnections.