The objective of this project is to scale up the applicability of planning algorithms developed in the field of Artificial Intelligence. The basic research component of the project is to first investigate techniques for improving scalability of planning using parallel techniques, then to explore machine learning techniques for reusing and refining generated plans. Industrial applications of AI planning algorithms have been limited largely by the computational complexity of the planning task. The significant speedup that can be obtained using parallel hardware and machine learning techniques can bridge the gap between research and a wide variety of industrial applications. To demonstrate the power of the techniques developed here, they will not only be formally verified but will be applied to such applications as automated assembly and cooperative planning of unmanned ground vehicles. The CAREER project will also impact graduate and undergraduate education. Existing seminar courses in Parallel AI and in Planning and Robotics will be refined. An industrial team will be formed to suggest topics and to oversee class projects. All course materials developed for these classes will be packaged and available for general dissemination over the Internet.
