Real-time planning and control of army UAVs under uncertainty

With advances in sensor technology and data fusion used in military operations, more information is available for decision making. A key question is how to make effective use of this information. Higher level sensors cue lower level sensors, in this case unmanned aerial vehicles (UAVs), to indicate potential target arrivals. Given probability distributions of these target arrivals, simulation and mathematical programming are used within a dynamic programming framework to determine control strategies for UAVs. An adaptive dynamic programming methodology is presented for the a uniform travel time UAV planning and control problem. Special structure of a network assignment problem is exploited to recursively update functional approximations representing future rewards through the network assignment problem's subgradient information. We develop an approximate dynamic approach to real-time planning and control of unmanned aerial vehicles with a focus on accounting for stochastic arrivals of new tasks. Experimentation demonstrates the use of this method and its potential for providing quick real-time controls for UAVs. Approaching the UAV routing problem using Adaptive Dynamic Programming offers a tractable framework in which to solve these difficult problems. Copyright © 2007 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.