Clemson Bionics – FoCI Spring 2017

Design and Development of Novel, Low-Cost Bionic Prosthetics

JL Robison, M Roach, A Sedler, C Hicks, A Ormerod, D McLeod, S Aitken, B Dorsey, M McCullough, J Rodriguez
Clemson University Department of Bioengineering, Clemson, SC
bionics@clemson.edu

Introduction

Problem Statement

In the United States, there are over 30,000 new below the elbow limb amputees each year [1]. Modern, electronically actuated prosthetics can provide restoration of form and function to those affected. Unfortunately, their high cost makes them unavailable to growing children, who require new sizes frequently, and to individuals in low-resource settings. Clemson Bionics is dedicated to the development of inexpensive, yet highly functional prosthetics that meet the needs of these amputees.

Aim

Our goal is to use versatile fabrication techniques and simple electronic systems to create bionic prosthetics with electronic control and feedback. We also serve the Clemson community by providing an opportunity for new members to learn and apply basic engineering principles and for senior members to develop leadership and teaching skills.

Challenges

During the design process, we have encountered several challenges and have developed strategies to move past them

  1. Affordable integration of electronic components for actuation and feedback
  2. Adapting to complex shapes without excessively complicated sensors and control algorithms
  3. Imperfect nature of fused deposition modeling (FDM) creates irregular joints that lead to mechanical instability

Methods

Main Goal 1: Robust underactuation

Underactuated systems have fewer actuators than degrees of freedom, and when used in combination with elastic elements they can allow for passive adaptability to objects of differing shapes and sizes [2]. Two competing designs were evaluated for potential use in the bionic hand. A pulley system and whippletree system were each created using a peg board, springs, and a stepper motor. We found that the pulley system was superior because it provided a mechanical advantage that will conserve power in the final device. We are currently in the process of designing a model that builds on these findings.

Main Goal 2: Mechanical design improvement for wrist

Our team has also been working to develop a strong and efficient wrist for our prosthetics. This required us to first define the necessary range of motion, then design a system which could achieve it. Through collaboration with a prosthetist from Greenville Health System, we determined that for most patients, only supination/pronation and flexion/extension were necessary, not adduction/abduction. From here we designed a wrist made of as few components as possible, resulting in a snaking rectangular shape that would provide both flexibility and rigidity.

Main Goal 3: Control system implementation

We are implementing the Myo electromyography (EMG) band and Raspberry Pi into our design to allow the user to control the prosthetic device using muscle contractions. Our team learned the basics of computing in Python with Raspberry Pi in order to interpret the output of the EMG band and generate output to control a DC motor based on these inputs. Ultimately, the EMG band will sense specific muscle contractions from the user, and output corresponding commands to linear actuators and DC motors that will control the device.

Main Goal 4: Integration of sensors and feedback

This part of the project involves the design and implementation of a haptic feedback system for the prosthetic. In our current design, force sensitive resistors at the tip of each finger provide a stimulus upon contact, which is translated to vibrational feedback via four eccentric rotating mass vibrating motors. Each motor corresponds to a single sensor and vibrates in various patterns and intensities depending upon the detected load of the resistor. Our group is currently exploring the design of a cuff containing four motors that will be worn on the bicep. These strategically placed motors will provide the user with increased spatial awareness regarding forces on the prosthetic.

Results and Conclusion

This semester, we have learned from our successes as well as our failures. We successfully recruited young bioengineering students, hosted informational talks on programming with the Arduino® platform, met with several groups of high school students, and made strides in many of our project areas. Our survey of amputee soccer players from Sierra Leone fell through, but this exposed members to the real challenges of research. We are working on another iteration of our base transradial bionic prosthesis prototype, based on shortcomings we identified internally. In the future, we will continue to refine the mechanical design using CAD software and hybrid deposition manufacturing (HDM) techniques, create a system of underactuating phalanges, implement an eight-channel electromyography (EMG) armband for control, and integrate sensors to provide haptic feedback. Overall, Clemson Bionics has been an excellent environment for students of all backgrounds to teach and to learn new skills while defining and addressing real needs.

Officers

President

Jonah Robison

jonahr@clemson.edu

Vice President

Christopher Hicks

cbhicks@clemson.edu

Chief Technology Officer

Andrew Sedler

asedler@clemson.edu

Treasurer

Matthew Roach

mcroach@clemson.edu

References

1. Amputee Coalition “Limb Loss Statistics” 2016
2. Ma, R et al “A modular, open-source 3D printed underactuated hand” 2013
3. Laliberte, T et al “Underactuation in robotic grasping hands” 2002
4. InMoov Robot by Gaël Langevin
5. e-NABLE Hand by e-NABLE Community
6. OpenBionics Project
7. Design of a Human Hand Prosthesis by Paul Ventimiglia
8. RealtimePlotter by Sebastian Nilsson
9. Ada Hand by OpenBionics