sA-ANT: A Hybrid Optimization Algorithm for Multirobot Coalition Formation

Abstract

Future critical missions will require intelligent systems that can autonomously form robust coalitions of heterogeneous robots. A novel hybrid coalition formation algorithm for spatially distributed real-world tasks in multirobot systems is presented, which combines ant colony optimization and a simulated annealing technique. The scalability of the algorithm is demonstrated by applying it to varying team sizes of up to 200 simulated heterogeneous robots. Experimental results show that the proposed algorithm generates better and higher quality coalitions when benchmarked against a heuristic coalition formation algorithm and two coalition formation algorithms that leverage an ant colony optimization approach.