Integrated control of a multiple-degree-of-freedom hand and arm using a reactive architecture based on high-speed proximity sensing

Abstract

We describe integrated control of a multiple-degree-of-freedom hand and arm using a simple reactive architecture based on high-speed proximity sensing. A proximity sensor installed on each fingertip of the hand detects the positions between the sensor and the surface of an object. By feeding back the sensor outputs, the reactive architecture simultaneously controls the position of the arm tip and the vector connecting the grasping position and the origin of the arm tip coordinates, the wrist posture, and the initial finger configurations before grasping. This architecture makes it possible to correct position errors. In addition, since the positions and postures are controlled without contact, there is no danger of damaging the object or the robot. To realize this integrated control using a simple reactive architecture, we designed a high-speed proximity sensor for the fingertips of the hand, and developed an integrated control system for the hand and the arm. We demonstrate high-speed position adjustment motion and grasping for static and moving objects in experiments.

Proximity Perception in Human-Centered Robotics: A Survey on Sensing Systems and Applications

“…In [65], [66], [67], Koyama et al developed finger-sized, high-speed proximity sensors mounting twelve photoreflectors on a fingertip as shown in Fig. 11 (a).…”

“…The authors also proposed a simple calibration method using changes in fingertip positioning and reflected light intensity. However, simple calibration methods, e. g. [58], [67], have relatively large errors (with millimeter or sub-millimeter accuracy) due to fingertip position errors or circuit noise. Therefore, reactive preshaping methods (see Sec.…”

“…Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. Koyama et al [65], [66], [67] realized independent control of joint poses of an 8-DoF robot hand (three fingers) using a high-speed optical proximity sensor. They demonstrated reactive preshaping for an apple, a banana, and moving objects on a conveyor or during handover, with simple joint-angle controls (see Fig.…”

“…17), updating the sensor value and control every 5 ms. In these preshaping controls [29], [67], the target value is set in advance from experimental data of an object set.…”

Review of Robot Skin: A Potential Enabler for Safe Collaboration, Immersive Teleoperation, and Affective Interaction of Future Collaborative Robots

“…In addition to the above body-covered proximity skin for human safety, finger-covered proximity skin can be used for secured grasping [98], [99]. As shown in Figure 5(e), Koyama et al [100] developed a high-speed proximity sensor for the robotic fingertips to control the positions and postures of robot hand without contact before grasping, which means there is no fear of damaging the end-effector or the object.…”