A Low-Complexity Autonomous 3D Localization Method for Unmanned Aerial Vehicles by Binocular Stereovision Technology

The unmanned aerial vehicles (UAV) boast for their small size, low cost, flexible maneuverability and good stealth, and are thus widely used in geological survey, target tracking and other fields. Without the expensive and bulky precision inertial navigation equipments, UAV could be difficult to localize during its course of flight because barely no global positioning system (GPS) signals can be detected. In order to address these issues, an autonomous three-dimensional (3D) localization method for UAV based on binocular stereovision technology is proposed in this conference paper. First, the optical principle of binocular stereovision technology and the estimation algorithm for the distance between the camera and the target are introduced. Then, both the least-squares (LS) and the least mean square (LMS) methods are applied in the localization algorithm. The complexities of both methods are calculated and compared. Simulation results validate that the proposed UAV autonomous 3D positioning method based on LMS has high positioning accuracy, strong universality and low complexity.