Parallel Stochastic Hill- Climbing with Small Teams

We address the basic problem of coordinating the actions of multiple robots that are working toward a common goal. This kind of problem is NP-hard, because in order to coordinate a system of n robots, it is in principle necessary to generate and evaluate a number of actions or plans that is exponential in n (assuming P 6= NP ). However, we suggest that many instances of coordination problems, despite the NP-hardness of the overall class of problems, do not in practice require exponential computation in order to arrive at good solutions. In such problems, it is not necessary to consider all possible actions of the n robots; instead an algorithm may restrict its attention to interactions within small teams, and still produce high-quality solutions. We use this insight in the development of a novel coordination algorithm that we call parallel stochastic hill-climbing with small teams, or Parish. This algorithm is designed specifically for use in multi-robot systems: it can run off-line or on-line, is easily distributed across multiple machines, and is efficient with regard to communication. We state and analyze the Parish algorithm present results from the implementation and application of the algorithm for a concrete problem: multi-robot pursuit-evasion. In this demanding domain, a team of robots must coordinate their actions so as to guarantee location of a skilled evader.