In recent times, NUI/NUX technology has found widespread application across a diverse range of fields, and is also utilized for controlling Unmanned Aerial Vehicles (UAVs). Natural control of UAVs requires the development of appropriate technology in advance. Path generation for natural UAV flight needs to be based on waypoints. UAVs normally fly straight from one waypoint to another. However, if flight between two waypoints is in a straight line, UAVs may collide with obstacles. In order to solve collision problems, flight records can be utilized to adjust the generated path taking the locations of the obstacles into consideration. This paper proposes a natural path generation method between waypoints based on flight records collected through UAVs flown by users. Bayesian probability is utilized to select paths most similar to the flight records to connect two waypoints. These paths are generated by selection of the center path corresponding to the highest Bayesian probability. While traditional approaches generated paths that led to UAV collisions, the proposed method generates paths that allow UAVs to avoid obstacles.