BINOCULAR CAMERA-BASED VISUAL SLAM METHOD FOR UNMANNED AERIAL VEHICLES, UNMANNED AERIAL VEHICLE, AND STORAGE MEDIUM

摘要

Disclosed in the present invention are a binocular camera-based visual SLAM method for unmanned aerial vehicles, an unmanned aerial vehicle, and a storage medium. The method comprises the steps of: obtaining depth images of at least two different locations by means of a binocular camera; obtaining camera pose information by means of a visual odometer according to the obtained depth images of the at least two different locations; performing nonlinear optimization, appearance-based loopback detection and loopback verification on the camera pose information to obtain optimized camera pose information; and performing binocular dense mapping according to the optimized camera pose information to obtain a global map. According to the present invention, depth images of different locations are obtained by means of a binocular camera, and a global map is obtained by performing binocular dense mapping upon the use of a visual odometer, nonlinear optimization, loopback detection and loopback verification. On one hand, the problem of interference caused by the use of RGB-D cameras can be resolved, and on the other hand, more precise localization can be achieved, and a more precise map can be built.