IJCAI-97 TUTORIALS

Practical Planning Systems

Steve Chien and Brian Drabble

Course Description

Automated planning is the generation of a low-level sequence of actions to achieve some desired world state while obeying domain constraints. Planning systems can be used to automate procedure generation problems and have been applied to such diverse fields as science data analysis, image processing, crisis response, space payload operations, and operating a network of communications antennas. Automated planning technology has the potential to reduce operations costs, decrease manual errors, and reduce dependency on key personnel.

This tutorial will cover the basic concepts in domain-independent artificial intelligence planning including: search, representing planning knowledge, plan and state space planning, operator- based planning and hierarchical task network planning. Advanced concepts such as planning and scheduling, decision theoretic planning, and mixed initiative planning will also be briefly discussed.

Important questions relevant to planning will be covered in the tutorial such as:

The tutorial will be broadly beneficial to a number of different groups, in particular:

Prerequisite Knowledge

Knowledge of basic concepts from Artificial Intelligence will be presumed : search, expert systems, logic-like representations. Familiarity with some planning and scheduling systems, basic search strategies, reactive systems, and/or scripting languages would be helpful but not essential.

About the Lecturers

Steve Chien is Technical Group Supervisor of the Artificial Intelligence Group, at the Jet Propulsion Laboratory, California Institute of Technology where he leads efforts in automated planning and scheduling for spacecraft mission planning, maintenance of space transportation systems, and Deep Space Network Antenna operations. He holds a B.S., M.S., and Ph.D. in Computer Science, all from the University of Illinois. Dr. Chien is also an Adjunct Assistant Professor in Computer Science at the University of Southern California.

Brian Drabble is a Research Associate at the Computational Intelligence Research Laboratory (CIRL) at the University of Oregon. His current responsibilities are to transition the planning and scheduling research being undertaken at CIRL into industry, commerce and military applications. Previous to joining CIRL he spent 8 years as a member of the Artificial Intelligence Applications Institute at the University of Edinburgh. His responsibilities included being project leader and co-principal investigator on the O-Plan project which is part of the $66 million DARPA/Rome Laboratory Planning and Scheduling Initiative. In addition he has worked with a number of clients including Toshiba, Hitachi, European Space Agency, and the British Government, to bring intelligent planning and scheduling into their organisations and products. Brian Drabble holds a B.Sc. from Staffordshire University and a Ph.D. from Aston University both in the U.K


higuchi@etl.go.jp
Last modified: Thu Feb 20 12:09:53 JST 1997