Temporal reasoning in plan management

Summary form only. While the traditional planning problem can be viewed as a temporal reasoning problem in the abstract, most planning algorithms use specialized constrained representations that are only useful for the planning task. In the TRIPS system, we need to build complex plans, but also to reason about them in different ways in order to modify plans, consider the effects of different plans over time, monitor execution, and support natural language interpretation. We call this range of reasoning tasks plan management, and it requires a more complex temporal representation. In fact, TRIPS uses a variety of temporal representations from the Interval Temporal Logic to concrete time-indexed representations. I describe the TRIPS system, examine the different forms of temporal representations used, and describe how each is appropriate for the different reasoning tasks it supports.