Abstract: With the development of deep space missions and the complexity of scientific tasks, the autonomous mission planning and scheduling of deep space explorers has become a research hotspot. On the basis of the task characteristics and the analysis of the system constraints of deep space probes,combined the intelligent planning theory with the constraint satisfaction technology,the dynamic characteristics of the constraints in the multi-layer constraint programming model is studied,and the fast extensional constraint filtering algorithm is designed based on the dynamic constraint sets. In this method, the newly added activities are classified according to the conflict between activities in the domain information,and the consistency of variables in the constraint table are checked. The results show that the proposed algorithm can effectively reduce the number of invalid constraints in the constraint processing,reduce the algorithm backtracking in the process of problem processing,and improve the efficiency and success rate of the planning.