The Multi-UAVs cooperative path planning has a complex search space in 3D environment, while the cooperative of Multi-UAVs is very hard to deal with. So a novel cultural algorithm based on spatial fuzzy set and differential evolution is proposed in this paper. The proposed approach uses fuzzy set to present spatial grid points in 3D, so that the key way-points on path should be paid more attention. Then the spatial fuzzy knowledge, history knowledge and cooperative knowledge that contained in the belief space of cultural algorithm prune the search space of Multi-UAVs path planning. Moreover, this approach uses differential evolution as the population space of cultural algorithm to generate the optimal solution, while it satisfies the constraints of multi-UAVs cooperative. The differential also extends the belief space with the unknown information of spatial to ensure the population diversity. In addition, the cultural algorithm exchanges the shared information, so that it accumulates the knowledge and revises the searching direction. The simulation results show that the spatial grid points based on fuzzy set enhance the efficiency of key way-points selected, and the cultural algorithm could explore more unknown space out of spatial grid such that it avoids the search fall into local optimization. Cooperative knowledge is also introduced to satisfy the requirements of Multi-UAVs cooperative to planning feasible paths for Multi-UAVs cooperative quickly.