This paper presents a path planning algorithm called D* lite-IPRM algorithm for Unmanned Aerial Vehicles (UAVs) in complex terrain.The operational environment consists of various shapes of obstacles,threatened areas and no-fly zones,which are usually called a joint name “obstacles” below to simplify.This algorithm can be divided into two steps,firstly Improved Probabilistic Roadmap Method (IPRM) is applied for local obstacle avoidance; secondly D* lite is used to plan a global trajectory.D* is an extension of A*,both of which are heuristic search algorithms.D* lite is a simplification of D* and at least of the same efficiency.It can repeatedly plan the optimal track from current point to the target.In our research,minimum flight distance is considered as the ultimate goal.We have discussed two-dimensional (2D) D* lite-IPRM and extended to 3D circumstance.This algorithm is demonstrated to be effective on both offline path planning and online replanning.