A Multilevel Learning Approach to Mobile Robot Path Planning

Abstract

The problem of adapting mobile robot navigation to changes in the environment is usually approached by modifying an internal world model. Descriptions on different levels of abstraction provide the information necessary for navigation and, therefore, influence the robot's behavior. The effect of such indirect adaptation is limited. The approach presented in this chapter describes a new technique for direct integration of navigation experience in path planning. Thus, not only the world knowledge, but also the planning behavior is improved over time. Experiments are carried out on a robot, which is controlled by a layered architecture. It is integrated in a multirobot control environment, which is described in the chapter. The focus of the chapter is toward improving the higher navigation levels. The main idea being presented is the realization of adaptive behavior not only on the level of reflexes, but also with respect to the planning capabilities of the robot. The application of learning techniques allows to continuously improve the estimation of plan costs and, therefore, the inherent strategy of the topological planner. It is illustrated that a combined learning of world description and navigation allows fast and sophisticated reaction to new environmental conditions.