A Grasp Strategy with Flexible Contacting for Multi-fingered Hand Rehabilitation Exoskeleton

To recovering the functions of hand after stroke, many hand exoskeletons and their control methods are developed. However, less research involves in the multi-fingered grasping. There are two primary problems: the fingers are correlative in the movement a

  • PDF / 514,196 Bytes
  • 13 Pages / 439.37 x 666.142 pts Page_size
  • 33 Downloads / 166 Views

DOWNLOAD

REPORT


Abstract To recovering the functions of hand after stroke, many hand exoskeletons and their control methods are developed. However, less research involves in the multi-fingered grasping. There are two primary problems: the fingers are correlative in the movement and the contacting part, the human finger, is flexible. This paper presents a method, which takes not only all the fingers but also their mechanical impedance into a dynamic system. The method is divided into three levels. First level, grasping planning, the desired interface force of each finger is derived by the geometric and external force information of object. Second level, multi-fingered coordinate force control, we see each finger’s impedance as a second-order subsystem to model an integrated coordinate dynamic system. Third level, single finger force control, execute the position and force command calculated in middle level by each finger, which has been presented in our early research. To verify the method, we set an experiment to grasp an apple assisted by a three fingers (thumb, index finger, and middle finger) exoskeleton. The results illustrate the effectiveness of the proposed method and also point out the direction for further research. Keywords Grasp

 Flexible  Coordinate  Multi-fingered  Exoskeleton

Q. Wei  C. Yang (&)  Q. Bi  W. Yang State Key Laboratory of Fluid Power Transmission and Control, Zhejiang University, Hangzhou 310027, China e-mail: [email protected] Q. Wei e-mail: [email protected] Q. Bi e-mail: [email protected] W. Yang e-mail: [email protected] © Zhejiang University Press and Springer Science+Business Media Singapore 2017 C. Yang et al. (eds.), Wearable Sensors and Robots, Lecture Notes in Electrical Engineering 399, DOI 10.1007/978-981-10-2404-7_18

225

226

Q. Wei et al.

1 Introduction Exoskeleton robots have been developed to replace the traditional therapistdependent post-stroke rehabilitation. Because the human hand is the most dexterous part of human body, including 5 fingers and 22 degrees of freedom (DOFs), development of hand exoskeletons has proceeded slowly in terms of both the mechanism and control method. Most of hand exoskeleton programs start from index finger exoskeleton then extend the method to other fingers. The common ultimate goal of them is recovering the function of hand. Thus the grasping with multi-fingered hand exoskeleton is studied widely. Here we classify the researches into three types, no interaction, interaction with virtual reality, and interaction with real object. The first type hand exoskeletons emphasize the range of movement (ROM) of each joint. Nakagawara et al. (2005) developed an encounter-type multi-fingered exoskeleton to control a dexterous robot hand. When slave hand touches an object, master finger is controlled to produce a force that is equal to the force applied on the slave finger (force control). The slave finger is always controlled to take the same position as that of the master finger (position control). This bilateral control is the so-called force-reflecting servo method. Fang et al.