Height estimation via stereo vision system for unmanned helicopter autonomous landing

Abstract

Auto landing is a very important function of control in the development of autonomous unmanned helicopters. In this paper, the stereo vision system is proposed to provide the height information on landing control instead of global positioning system (GPS). GPS provided 3-axis information, but the height error of GPS is usually in meters, which is not accurate enough for landing. The stereo vision system provided a more accurate height information to correct the error of GPS. In this paper, the stereo vision system is composed of two webcams and the variable baseline method is selected to increase the measurement range. Experimental results show that variable baseline method did increase the accuracy of the stereo vision system with the webcams, and that not only provided an authentic piece of height information but also provided another option of sensor for autonomous unmanned helicopter's auto landing control. © 2010 IEEE.