基于流演算的智能体行动推理模型设计与实现

Design and implementation of reasoning about action of intelligent agent model based on fluent calculus

并发约束程序设计在人工智能程序设计领域中占据越来越重要的位置,约束处理规则作为新一代的并发程序设计正倍受关注。对约束处理规则和流演算理论及其实现语言FLUX进行了研究,结合流演算和JCHR推理模型优点,设计了一种基于Java的流演算解释器JFLUX,同时提出了一个基于目标驱动的,在不完全可知的虚拟环境中通过感知到的有限信息进行自主行动推理能力的智能体模型,实现了办公室场景中智能体行动推理系统。

Concurrent constraint programming（CCP） has become more and more important in the field of artificial intelligence（AI）.As the new generation of CCP, constraint handling rules（CHRs） have recently been drawn considerable attention.The JFLUX interpreter using Java language and the target-driven intelligent agent model having autonomous behaviors ability are introduced, and they are achieved by combining the advantage of the JCHR reasoning model and fluent calculus, based on the research on the theory of CHRs and fluent calculus and its implementation language flux.In addition, the intelligent agent can reason about action automatically using the limited information acquired from sensor in an incomplete, dynamic environment for any given goal.Finally, action planning system for the intelligent agent of office scenario is achieved.