Mission planning and scheduling of satellite, which is to support task plan making of satellite scientifically, is a main content of space resource management. One of the most important characteristics of satellite mission scheduling is that it is subject to time windows constraint, which means that tasks can only be completed within the given time windows. The scheduling model is formulated based on reasonable assumptions and constraint programming method. The model is solved with Variable Neighborhood Tabu Search (VNTS) algorithm. On the basis of a limited experiment, we observe that the algorithm is very effective in obtaining near-optimal solutions.