Case-based path planning for autonomous underwater vehicles

Case-based reasoning relies on past experience to solve new problems. A new solution is generated by retrieving and adapting an old one which approximately matches the current situation. Here, the authors outline a case-based reasoning approach to path planning for AUV missions. An annotated map system is employed to represent the navigational environment. Previously used routes, represented as objects in the map, are retrieved and repaired to arrive at a plan for a new route. Whenever a matching route is not available, a new route is synthesized by the planner relying on past cases that describe similar navigational environments. The case-based approach is used not only for adapting old routes but also for synthesizing new ones. Thus, the planner will not fail just because an appropriate matching case is not available. In this paper, the authors describe the representation of the environment including past routes and objects in the navigational space and the representation of case frames of past route planning scenarios. Further, the authors present the retrieval and repair strategies and the scheme for generating new routes. This case-based path planner is part of a larger planning system for the authors´ AUV covering all aspects of missions such as navigation, task execution, resource allocation, and safety safe-guard