Probabilistic State-Dependent Grammars for Plan Recognition

David V. Pynadath and Michael P. Wellman

Techniques for plan recognition under uncertainty require a stochastic model of the plan-generation process. We introduce probabilistic state-dependent grammars (PSDGs) to represent an agent's plan-generation process. The PSDG language model extends probabilistic context-free grammars (PCFGs) by allowing production probabilities to depend on an explicit model of the planning agent's internal and external state. Given a PSDG description of the plan-generation process, we can then use inference algorithms that exploit the particular independence properties of the PSDG language to efficiently answer plan-recognition queries. The combination of the PSDG language model and inference algorithms extends the range of plan-recognition domains for which practical probabilistic inference is possible, as illustrated by applications in traffic monitoring and air combat.

This page will eventually have more information about this work. Right now, it serves as an appendix to the following paper:

The following PS files provide pseudocode for the algorithms described in the corresponding sections of the paper:

Bear in mind that these files are under construction to remove inconsistencies with notation from the paper and other formatting.