Intelligent planning is an important branch of artificial intelligence, which mainly related to the arrangements and achieve about sequence of actions or strategies. It's important in many application areas. Conformant planning is the planning in uncertainty about the initial state and action effects. It greatly extends the planning application in the real capacity and make planner is more useful for actual problems. Presently, the main idea of most state-of-the-art conformant planners is to transform a conformant planning problem into a search problem in the space of brief states. This method greatly improved the efficiency, but only gets the non-parallel planning solution.In this paper, we introduce the new definition of independent action and interaction, given the rules to generate parallel action in conformant planning. We design a system called CFFP on the basis of an efficient conformant planner CFF (Conformant Fast-Forward). CFFP generates parallel plans by using recursion online parallelization of partial plans. And modify the Enforced Hill-climbing and"Helpful Actions"of CFF to help generate parallel planning. This method makes the belief state-space search algorithm can generate parallel planning, finds a balance between search efficiency and solution quality.We have developed an online parallelization planning system CFFP by C programming language based on the CFF. The system is capable of solving the conformant planning problems and classical planning problems, find the less time-step approximate optimal solution, and improve the quality of planning solutions. The empirical results show that it can generate good plan close to the optimal parallel planning by small cost compared to CFF. |