Unit 5: Automated Planning and Acting

Commentary

This unit deals with planning: one of the classic tasks whose aim is to produce a sequence of actions for achieving a goal. You may have seen many automated planning systems based on AI techniques in recent years; for example, Remote Agent - the first autonomous planner-scheduler to control a spacecraft when it flew on board the Deep Space One probe. This unit focuses on the basic concepts, representations, and algorithms of planning, such as planning with state-space search, partial-order planning, and the planning graph. Planning and acting tasks in the real world, such as scheduling, conditional planning, and multiagent planning, will then be introduced to explore some complex planning tasks.

Unit Purpose

When you complete this unit, you will be able to

• Discuss basic planning problems and their representation.
• Describe the state-space search techniques used to solve planning problems.
• Explain the principles of partial-order planning and graph-plan algorithms.
• Discuss time and resource constraints in real-world planning problems.
• Describe HTN planning, conditional planning, and multiagent planning problems and mechanisms.

Section 1: Planning problems and algorithms
Section 2: Planning with time and resource constraints
Section 3: Conditional, continuous, and multiagent planning

Readings

Supplemental Unit Readings

Books: Ghallab, M., Nau, D., and Traverso, P.. (2004). Automated planning: Theory & practice. Morgan Kaufmann.

Activities

• Explore other research and application events regarding planning and scheduling, especially some systems in aerospace and e-learning.
• Share your findings from the above activity through the online course conference.