A Probabilistic Framework for Passive Cooperation Among UAV’s Performing a Search

This paper presents a probabilistic framework for short-term (within a limited planning horizon) passive cooperation among unmanned aerial vehicles (UAV’s) performing a search. This structure is suitable for solution using a Dynamic Programming (DP) planning algorithm. Because of the exponential computational complexity of the DP algorithm, an ideal cooperation scheme is not feasible for implementation, so an approximation of the ideal case must be used. There exists a fundamental trade-off between accuracy and complexity when using this approximation. Extending significantly previous work, the framework presented in this paper provides flexible methods for adjustment of this trade-off, specifically in making it more accurate, if necessary. This takes the form of explicitly accounting for multiple vehicle interference in the one-step gain function of the planning algorithm, and by approximating the probability of vehicle interference using dynamic environment information. Some simulation results are shown to demonstrate the effectiveness of the approach.