Abstract
A pursuit problem is a multi-agents' benchmark problem, where four hunters pursue and capture the prey in a cell environment. We have proposed a staged view that reflects a human's view where we can see an object easily in the neighborhood but more difficult in the longer distance and easily in the center direction but more difficult in the righter and lefter directions in both cell environment and real number environment. We showed that hunters can capture the prey even with the staged view. In the previous research, if hunters lose the prey, they select no special actions. However, human will try to surmise the prey's position. In this paper, we apply a fuzzy Q-learning to hunter's learning. Hunters surmise prey’s lost position from the previous prey's position and their own action by fuzzy Q-learining. We simulate this problem, and have a result better than the previous one.
