UAV Trajectory Planning for Complex Open Storage Environments Based on an Improved RRT Algorithm

Abstract

Multi-rotor UAVs (Unmanned Aerial Vehicles) have been increasingly used for hazardous inspection tasks in complex open-air warehouse storage environments due to their high maneuverability and aerial perspective. To facilitate rapid response to patrol missions and improve the efficiency of UAV trajectory planning. This paper established a rotary-wing UAV trajectory plan model considering UAV patrol efficiency, trajectory cost, and power consumption cost. Secondly, an improved SSA (salp swarm algorithm) is incorporated for the shortcomings of low algorithmic search efficiency and unsmooth paths when planning paths in the traditional RRT (Rapidly-exploring Random Trees). The predation mechanism of the salps group is incorporated into the random sampling of the RRT algorithm, which reduces the invalid sampling of random points and introduces the adaptive leader structure, and reverses the search strategy to improve the algorithm’s global search for superiority at the later stage of the search. Finally, the designed LASSA-RRT algorithm is subjected to simulation experiments and compared with RRT, RRT*, IRRT, and PF-RRT* in a cross-sectional manner. The results show that the LASSA-RRT algorithm has an average reduction of 55.83% in sampling times, 51.91% in run time, 13.17% in track length, and 0.1491% in flight cost. In summary, this paper’s UAV trajectory planning method can be effectively applied to complex open storage environments. It can provide a helpful reference direction for UAV trajectory planning.

Keywords

• RRT
• salp swarm algorithm
• path planning
• UAV
• adaptive
• reverse search