A cluster first strategy for distributed multi-robot task allocation problem with time constraints

This paper considers the problem of having a team of mobile robots to visit a set of survivors before their specified deadlines in a distributed multi-robot system. The objectives are to maximize the number of rescued survivors, minimize the waiting time of survivors, and minimize the total traveling distance of robots, with priority from high to low. In this paper, a cluster-first strategy is proposed, which builds upon existing consensus-based distributed task allocation algorithms. The basic idea is that assigning a group of tasks that are clustered to each other to a robot is more likely to result in an efficient schedule, thus reducing the waiting time for survivors and the robots traveling distance, which in turn may increase the number of people rescued. Each robot updates its cluster at the beginning of each iteration and prioritizes adding survivors in the cluster to its rescue plan. Simulated rescue scenarios are used to evaluate the performance of the proposed strategy. The proposed strategy is applied to two representative consensus-based distributed algorithm, the consensus-based bundle algorithm (CBBA) and the performance impact (PI) algorithm. Experimental results show the applicability and the outstanding performance of improving solution quality of the proposed strategy.