Developing a General Contingency Planner, Phase 2

Abstract : This report summarizes the work performed in the final phase of a 3- year effort to investigate basic mechanisms for solving planning problems in domains characterized by adversity. In the first phase of the effort, a general purpose planning mechanism was developed. This planner represented plans in a manner content with the goal tree formalism characteristic of Artificial Intelligence action planning research and used plan generation/search techniques derived from both AI action planning and knowledge-based game-playing theory. In the second phase of the effort, advanced planning mechanism was also extended from its initial domain (the two-player board game of Othello) to an Army maneuver planning domain.