An Inert and Efficient Fusion Replanning Method for UAV Fast Autonomous Flight

Abstract

Path planning is one of the key components to achieve fast autonomous flight of unmanned aerial vehicle (UAV) in unknown environments. However, it remains a challenge to generate a high-quality trajectory with less time cost and less replanning number. In this paper, an inert and efficient fusion replanning method is proposed. At first, we design a fast and robust guiding path generation method. Based on the guiding path, we build the safe flight corridor (SFC) and generate the high-quality initial flight trajectory by using the hard-constrained method. And then, to quickly avoid newly discovered obstacles, the gradient-based method is used to generate a local new collision-free trajectory in real-time based on the initial flight trajectory. Finally, we design an inert replanning strategy to reduce the number of replanning on the premise of ensureing the trajectory quality. Benchmark comparisions and real-world experiments are conducted to verify the efficiency. Experimental results show that the proposed method achieved better performance compared with other methods.