Control of Small Unmanned Aircraft Based on Embedded System and Global Visual Navigation

Abstract

Small unmanned aircraft has been the subject of extensive research because of its wide use in civil and military fields. In this paper, we have set up a global visual navigation platform and completed the autonomous flight control of the aircraft. On measurement aspect, locations and attitude angle are obtained by image processing algorithm. The paper describes an adaptive image segmentation algorithm based on color space, which can make accurate detection in different light conditions. Camera correction algorithm is applied to provide locations with higher precision, and the feedback rate of image program meets system requirements well. On control aspect, digital control of the system is decoupled into four channels. We adopt an incremental PID controller for each channel and achieve good response effects respectively. The aircraft performs well in hovering and autonomous flying according to planed path, which demonstrates the stability and accuracy of the system.