Research on UAV's Autonomous Target Landing with Image and GPS under Complex Environment

Autonomous target landing of unmanned aerial vehicles (UAVs) is widely used in many fields. At present, most of the solutions have two main problems that need to be improved. First, the landing accuracy is low; second, the required sensors are complicated and high-cost. In this study, improvements are made to these problems respectively. For the former problem, an image processing algorithm based on Relative Total Variation (RTV) model is used to find landmarks. For the latter problem, a bi-objective optimization model is proposed to fuse GPS and image processing navigation results, and a landing algorithm is also proposed by using simple and low-cost sensors. A comprehensive autonomous target landing algorithm can be obtained by combining two improvements.