Getting a Grip: Development and Experimental Validation of a Control Framework for Precise and Dexterous Robotic Hand Manipulation
TEX219
Dapeng Li
5089106872
dli@umassd.edu
https://umassd.zoom.us/j/93841383671?pwd=2NKzi6vAtyoyyiym03dWNbPV4JqVag.1
Abstract:
Tendon-driven robotic and prosthetic hands face several mechanical and control challenges, including maintaining consistent tendon tension, balancing motor torque, managing cables, achieving precise movement, and preventing mechanical failure. Furthermore, incorporating biological signals, such as electromyography (EMG), for intuitive prosthetic control presents further challenges. The purpose of this project is to design and develop a fully electric robotic hand and control framework that can translate a variety of input signals into precise and responsive movements. The system will be designed to provide consistent control of the robotic hand which will establish a platform that can ultimately support biological control inputs such as EMG. This system’s performance will be evaluated based on movement precision, responsiveness, repeatability, and the ability to produce controlled movement using different types of input signals. The robotic hand will integrate motors, encoders, microcontrollers, tendon-driven mechanisms, and geared transmission systems within a mechanically optimized design developed using computer-aided design and additive manufacturing. Following construction and integration of the robotic hand, experimental testing will characterize its mechanical and control performance and identify limitations affecting precise and dexterous movement. The resulting control framework and experimental data will establish a foundation for continued research into robotic and prosthetic hand control. This includes the future integration and evaluation of biological control signals. The system will additionally provide a platform for subsequent experimentation, academic research, and development of advanced prosthetic control strategies.