BACO: A Bi-Ant-Colony-Based Strategy for UAV Trajectory Planning with Obstacle Avoidance

Abstract

Trajectory planning for a logistic delivery using an unmanned aerial vehicle (UAV) involves a typical traveling salesman problem (TSP), in which the turning of the UAV to avoid obstacles can cause significant energy consumption. The obstacles in the airspace and the angle constraints of the UAV must also be considered in the delivery. To address the low precision of UAV trajectory searches, and the serious impact of flight angles on UAV energy consumption, we propose a UAV trajectory planning strategy called bi-ant-colony optimization (BACO). BACO consists of two phases: path planning and track planning. By applying the guidance layer ant colony optimization (GuLACO) algorithm, the path planning phase eliminates the problem of ant colony deadlock that arises in multi-target point environments, and reopens the ant tabu table to search for a guidance path. Following this, the track planning phase employs the general layer ant colony optimization (GeLACO) algorithm to build the guidance path in segments. Furthermore, the precision of the flight heading for the UAV is optimized by adjusting the flight step in an adaptive manner, and obtaining fine-grained UAV flight tracks to control the turning angle of the logistics UAV. Our simulation results show that compared with the use of the greedy algorithm and the classical ACO algorithm, UAV trajectory planning using BACO can not only obtain shorter flight paths that take into account obstacle avoidance, but can also reduce the energy consumption of the UAV by finely controlling the amplitudes of the flight angles to ensure the safety and energy efficiency of UAV while in flight.