Laser-based Collision Avoidance and Reactive Navigation using RRT* and Signed Distance Field for Multirotor UAVs

摘要

Collision avoidance plays a crucial role for autonomous navigation in unknown dynamic environments and still remains an ongoing research problem. In this paper, we present a new collision avoidance algorithm by combining an RRT* path planner with a Signed Distance Field (SDF) based collision checking algorithm, in which the trajectory is optimized by a short cut and Optimal Polynomial Trajectory algorithms. The proposed algorithm is integrated to work in combination with a Model Predictive Control (MPC) based trajectory controller in order to provide a complete system for reactive navigation purposes. A thorough evaluation of the proposed algorithm has been conducted in several simulating scenarios using RotorS Gazebo simulator, showing fast collision checking capabilities in the presence of static and dynamic obstacles. The results show that the proposed algorithm outperforms in 76.93% considering the processing time when tested in a 1000 × 1000 pixels map. The results also demonstrate that the proposed navigation algorithm allows the safe navigation of a multirotor Unmanned Aerial Vehicle (UAV).