In this work we introduce a real-time path planning method for Unmanned Aerial Vehicles (UAVs) in uncertain environment. The kinematic system of the vehicle is therefore heuristically reduced to a set of feasible trim trajectories. The local operating Iterative Step Method (ISM) sequentially determines the next best trim trajectory, which minimizes a local cost-function. By varying the respective weighting factors, it’s possible to get a trade off between path length and risk of the path. The local ISM is then integrated in a graph-based search of the best global by using the algorithm of Dijkstra. Finally a procedure for the real time use of the global path planner is presented.