Robotic Arm for Mimicking Human Arm Movements
Location
Atrium
Session Format
Poster Presentation
Research Area Topic:
Engineering and Material Sciences - Electrical
Co-Presenters and Faculty Mentors or Advisors
Imani Augusma
Dr. Fernando Rios-Gutierrez
Abstract
Robotic mimicking has begun to be a new important topic of research. Robotic mimicking is a convenient way to teach robots the operations they are to perform. The advantage of mimicking is that instead of developing a complex control system, a human can demonstrate the operations that a robot needs to perform so that the robot could mimic those operations. In this project, we propose to design and implement a robotic arm with 3 DOF that is capable of mimicking simple human arm actions. The implementation of this project involves three major modules: a) The capture of signals from the human arms, in particular the shoulder, elbow, and wrist using wireless motion tracking sensors, b) the classification of the signal to identify the particular movement being performed, and c) the electronic interface to control a robotic arm that will mimic the movements.
This will be implemented by using XSENS wireless motion trackers. A library of arm motions will be created using a variety of test subjects to provide a vast cross-section of data to use. Using the information taken from these sensors, a neural network will be created to recognize the specific motions that are to be mimicked by the robot arm. From there, the outputs of the neural network will be programmed to a robot arm with 3 DOF.
Keywords
EMG, 3 DOF, Robotics, Wireless motion tracker, Neural network, Robotic mimicking
Presentation Type and Release Option
Presentation (Open Access)
Start Date
4-24-2015 10:45 AM
End Date
4-24-2015 12:00 PM
Recommended Citation
Stroud, Joshua L., "Robotic Arm for Mimicking Human Arm Movements" (2015). GS4 Georgia Southern Student Scholars Symposium. 36.
https://digitalcommons.georgiasouthern.edu/research_symposium/2015/2015/36
Robotic Arm for Mimicking Human Arm Movements
Atrium
Robotic mimicking has begun to be a new important topic of research. Robotic mimicking is a convenient way to teach robots the operations they are to perform. The advantage of mimicking is that instead of developing a complex control system, a human can demonstrate the operations that a robot needs to perform so that the robot could mimic those operations. In this project, we propose to design and implement a robotic arm with 3 DOF that is capable of mimicking simple human arm actions. The implementation of this project involves three major modules: a) The capture of signals from the human arms, in particular the shoulder, elbow, and wrist using wireless motion tracking sensors, b) the classification of the signal to identify the particular movement being performed, and c) the electronic interface to control a robotic arm that will mimic the movements.
This will be implemented by using XSENS wireless motion trackers. A library of arm motions will be created using a variety of test subjects to provide a vast cross-section of data to use. Using the information taken from these sensors, a neural network will be created to recognize the specific motions that are to be mimicked by the robot arm. From there, the outputs of the neural network will be programmed to a robot arm with 3 DOF.
