Probabilistic State-Dependent Grammars for Plan Recognition
This work addresses plan recognition for domains requiring practical probabilistic inference, though it appears incremental as it builds on existing grammar-based methods.
The paper tackled the problem of plan recognition under uncertainty by introducing Probabilistic State-Dependent Grammars (PSDGs) to model an agent's plan-generation process, extending probabilistic context-free grammars to allow state-dependent probabilities, and demonstrated its application in domains like traffic monitoring and air combat.
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.