Multiagent Real-Time-A* with Selection: Introducing Competition in Cooperative Search

A new cooperative search algorithm that introduces a GA-like selection mechanism is developed. In this algorithm, a state-space search problem is solved concurrently by multiple agents, each of which executes the Real-Time-A* algorithm. These agents compete for existence, i.e., each agent periodically reproduces o spring stochastically based on its tness de ned by the heuristic estimation value of its current state, so that an agent in a good state tends to reproduce many o spring while an agent in a bad state tends to be exterminated. Experimental evaluations show that this algorithm is very e ective for problems that can be divided into serializable subgoals (e.g., n-puzzles), although the agents do not have any knowledge about these subgoals. In particular, this algorithm can solve the 48puzzle, which can not be solved by existing heuristic search algorithms consistently within a reasonable amount of time unless the knowledge about the subgoals is explicitly given.