Multitarget Pursuit Coalition Game of Radar-Enabled UAVs

This paper explores the problem of tactical unmanned aerial vehicles (UAVs) equipped with radar systems for pursuing multiple spatially-separated noncooperative targets. The multitarget pursuit problem of radar-enabled UAVs (RUAVs) is formulated as a multiagent cooperative tracking and control process and as a distributed constrained optimization problem considering the costs of collision, threat, and flight distance as well as kinematic constraints. To solve this problem, each RUAV as an autonomous agent uses the radar to actively sense the states of targets and utilizes the sensing results for task assignment and path planning with other RUAVs cooperatively to catch up with all targets as quickly as possible. This coordination behavior of RUAVs is simulated as the foraging behavior of the bee colony and hence the artificial bee colony- like RUAV swarm system based on the coalition game with transferrable utility is developed. Then, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">distributed Nash equilibrium seeking multitarget pursuit (DNESMP)</i> algorithm consisting of target state prediction, coalition formation, path planning, and target state estimation is proposed to find the Nash-stable coalition partition. Numerical results illustrate the effectiveness and superiority of the DNESMP algorithm over typical counterparts in terms of task completion time and computation consumption.