Autonomous Control of Micro Flying Robot

Abstract

In the case of a natural disaster such as an earthquake, a micro air vehicle (MAV) would be very effective for surveying the site and environment in dangerous areas or narrow spaces that humans cannot access safely. In addition, it would be helpful for preventing secondary disaster. This paper discusses autonomous hovering control and guidance control of a micro flying robot ( μFR). Since a μFR is very small and light, sensors such as a global positioning system (GPS) cannot be carried. So, the three-dimensional position of a μFR is measured by using one charge coupled device (CCD) camera to recognize a marker attached to the μFR. We adopted proportional—integral—derivative (PID) control because its performance can be increased by tuning, even when there is no model. Nevertheless, it is not easy to find the gain of the PID controller for the X and Y directions. Therefore, the first step that we took is modeling based on system identification and then we designed a model-based controller.