Real-Time Multiple Points Tracking and Localization Using Wide Angle Cameras for Indoor Flying Robot

Abstract

Using other visual information of object to compute its location is attracting lots of researchers and engineers in recent years. In this pager, a new real-time multiple points tracking and localization system is introduced while keeping the robustness in a certain degree. The ma¬j¬or contributions of our paper are: several infrared LEDs are used to avoid the interference of visible light, some wide-angle lens of the cameras are incorp¬orated to expand th¬e v¬isual field of cameras, and a real-time multiple points t¬r¬acking method is used to accelerate the points association among several ima¬g¬es captured from different cameras. In experiments, we mounted the points o¬n the flying robot to demonstrate that the real-time tracking is robust and the localization accuracy meets the application requirements.