Development of a Multifingered Hand With Proximity Sensors for Aerial Manipulation

Abstract

In aerial manipulation, obtaining sufficient information near an object is difficult because visual sensors occlude and interfere with grasping. To address this issue, in this study, a robotic hand that uses proximity sensors is proposed for aerial manipulation. These sensors can detect an object without contact or occlusion when the object is close. The robotic hand developed in this study has a palm and fingertip proximity sensors. In proximal areas, the position of an unmanned aerial vehicle (UAV) is controlled using information from the palm sensor to reach the center of the object. After approach motion, the fingers of the hand are controlled based on the information from the fingertip sensors to maintain the relative position between the fingertips and object before grasping (hereafter, called pregrasp control). This control method enables rapid and robust object grasping. The experiment on simultaneous UAV position control by the palm and fingertip position control by the fingertip sensors demonstrated hovering within <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$16.8~\pm ~19.5$ </tex-math></inline-formula> mm, subsequently stable pregrasp control, and finally successful object grasping. In conclusion, the proposed approach using the multifingered hand with proximity sensors is effective for aerial manipulation.