Perching with a robotic insect using adaptive tracking control and iterative learning control

摘要

Inspired by the aerial prowess of flying insects, we demonstrate that their robotic counterpart, an insect-scale flapping-wing robot, can mimic an aggressive maneuver seen in natural fliers-landing on a vertical wall. Such acrobatic movement differs from simple lateral maneuvers or hover, and therefore requires additional considerations in the control strategy. In this paper, we propose a single-loop adaptive tracking flight control suite designed with an emphasis on the ability to track dynamic trajectories, and an iterative learning control algorithm to account for unmodeled dynamics and systematic errors for improved landing accuracy. Magnets were chosen to enable attachment to the vertical surface due to their simplicity. The proposed controller was verified in a series of hovering and aggressive translational flights. Furthermore, we show that by learning from previous failed attempts, the biologically-inspired robot could successfully perch on a magnetic wall after eight learning iterations.