Probabilistic BDI Agents: Actions, Plans, and Intentions

The Belief-Desire-Intention (BDI) architecture is a popular framework for rational agents, yet most verification approaches are limited to analysing qualitative properties, example whether an intention completes. BDI-based systems, however, operate in uncertain environments with dynamic behaviours: we may need quantitative analysis establish properties such as the probability of eventually completing intention. We define probabilistic extension Conceptual Agent Notation (CAN) BDI agents that supports action outcomes, and plan selection. semantics executable via encoding Milner’s bigraphs BigraphER tool. Quantitative conducted using PRISM. While new can be applied any CAN program, demonstrate by comparing standard selection strategies (e.g. ordered or fixed schedules) evaluating executions smart manufacturing scenario. results show improve significantly completion, appropriate distribution. also impact outcomes marginal, even when failure probabilities large, due agent making smarter choices.