Y. Meng and K. Cao (USA)
Multi-robot system, target searching, decision making, game theory.
A multi-robot system is applied to search for a target in a partially known dynamic environment, where the searching area is partitioned into smaller regions and a priori probability of target distribution for each region is given. A strategic game approach is proposed to optimize the searching task using a dynamic-programming-based utility function. The pure Nash Equilibrium solution and the mixed-strategy equilibrium solutions are provided. Extensive simulation results demonstrate that, compared to the other heuristic search strategies, the game theory not only consumes less searching time, but also be more robustness to handle the environmental uncertainty. Since the game theory based coordination obviates the explicit communication, it can be applied to some emergency scenarios, such as in the urbane search and rescue, where the explicit RF communication tends to attenuate.
Important Links:
Go Back
