[For the entire collection see Zbl 0707.68001.] Starting from a criticism of the previous approaches on intelligent agent modelling for failing to provide a formal analysis of the defeasible reasoning inherent in plan recognition, the authors develop an alternative formalization of plan recognition as belief and intention ascription. Maintaining much of the complexity and subtlety of the conflicts coming from the possible ascriptions in the plan recognition process, the approach gives some more important general principles of conflict resolution. It is conjectured that plan recognition is a “bottom-up” recognition process, with global coherence used mostly as an evaluative measure, to eliminate ambiguous plan fragments emerging from local cues. This preliminary report concentrates on the formal definition of local ascriptions, along with their interactions. The direct argumentation system ARGH is used to incrementally allow explicit statements about the resolution of conflicts between candidate ascriptions.
Reviewer: N.Curteanu