Adapting to Changing Goals
Almost all existing heuristic search algorithms assume that the goal state is fixed and does not change during the course of the search. For example, in the problem of a robot navigating from its current location to a desired goal location, it is assumed that the goal location remains stationary. In this chapter, we relax this assumption, and allow the goal to change during the search [Ishida and Korf, 1991; Ishida 1992, Ishida and Korf 1995]. In the robot example, instead of moving to a particular fixed location, the robot’s task may be to reach another robot which is in fact moving as well. The target robot may be cooperatively trying to reach problem solving robot, it may be actively avoiding the problem solving robot, or it may be indifferent and simply going about other business. There is no assumption that the target robot will eventually stop, but the goal is achieved when the position of the problem solving robot and the position of the target robot coincide. In order to guarantee success in this task, the problem solver must be able to move faster than the target. Otherwise, the target could evade the problem solver indefinitely, even in a finite problem space, merely by avoiding being trapped in a dead-end path.
KeywordsProblem Solver Problem Space Heuristic Function Optimal Move Performance Bottleneck
Unable to display preview. Download preview PDF.