Traditional global path planning algorithms include artificial potential field method, genetic algorithm, intelligent bionics algorithm, heuristic algorithm and so on. However, these methods all need to model obstacles in the known global space, and are not suitable for solving the planning problem of multi-degree-of-freedom robots in complex environments. The path planning algorithm based on rapidly exploring random tree, through the collision monitoring of sampling points in the state space, avoids the modeling of the global space, and can effectively solve the path planning problems of high-dimensional space and complex constraints. By comparing with the artificial potential field method and the A*algorithm, the advantages of the RRT algorithm in solving the UAV path planning problem in a complex environment is determined in the paper. After optimizing the relevant parameters, the method is probabilistic and has an optimal solution, and applied in the Fixed-wing intelligent cluster flight formation control and coordination project at the same time.