Tracking a dynamic invading target by UAV in oilfield inspection via an improved bat algorithm

Abstract

A novel dynamic invading target tracking method for the oilfield inspection by unmanned aerial vehicle (UAV) is presented in this paper. In this study the quad-rotor UAV is used to track an invading target, because the traditional manual inspection method and fixed-points video monitoring method has some drawbacks such as low efficiency, high cost, blind spot, and so on. A trajectory prediction method for the ground dynamic invading target is firstly proposed to predict the moving trajectory of the invading target. Then, the swarm intelligence based optimization algorithm is used to optimize the tracking trajectory of UAV, which in order to keep the distance between the UAV and the target closing to the desired distance during tracking process. In order to overcome some drawbacks such as easily being fallen into the local optimal solution and poor stability of the optimization, an improved bat algorithm (named FOBA) is proposed to improve the local searching ability of the bat algorithm (BA), which uses a food searching mechanism in the fruit fly optimization algorithm (FOA). Case studies are conducted with the desired distance is 50m between the UAV and the target, and experimental results show that the FOBA algorithm can effectively keep the tracking distance between the UAV and the target being about 55m, which is better than some other methods.