Biomimetic measurement of optical flow and centroid for visual-servo control of hover flight

This paper presents a novel, biomimetic design of a visual sensor and visual servo system for a micro-RUAV, based on a study of the hybrid controls approach used by Macroglossum Stel. L. for pure 2-D hover stabilization. The sensor approach determines ego-motion in the Yaw axis through use of an optic flow estimator on a peripheral ring about a foveal region centered on a target. A centroid measure of the target is used to govern remaining degrees of freedom. The biomimetic controller builds upon locomotor studies and uses a PID control to govern error corrections for each separate axis. In turn, the corrective motions of each axis are combined through a thrust matrix, which projects the desired thrust on a series of pre-learned signals in cardinal directions about the RUAV. Experimental implementation and results are discussed.