Efficient local path planning for UAVs in unknown environments

摘要

This paper presents a local path planning method for real time applications with Unmanned Aerial Vehicles in unknown environments by considering an onboard sen- sor of a limited range. Both static and mobile obstacles are considered. The collisions are detected from a laser sensor readings and an efficient algorithm computes the solution trajectory to avoid them. The proposed ap- proach generates a visibility graph and then a Dijkstra's algorithm is run to compute the shortest path. The execution time obtained is low because the proposed approach adds few nodes to model both static and mobile obstacles. The algorithm has been integrated in ROS (Robot Operating System) framework. Many simulations have been performed in different scenarios with a dynamic quad-rotor model to test the proposed algorithm.