I was a research assistant in the Bretl Research Group at UIUC from 2013-2016. During my first year (2013-2014) I worked on haptics in prosthetics, more specifically proprioception for myoelectric hands. I was fortunate enough to co-author a paper in the Eurohaptics 2014 conference. I stayed to work in the lab full time during the summer of 2014 and continued this research while simultaneously helping lead a team of undergrads to make a complete low-cost myoelectric hand. A graduate student, Aadeel, and myself started a collaboration with a non-profit prosthetics group, ROMP, based in Ecuador. We received funding from the US Embassy and sent our low-cost myoelectric hand to be implemented on an actual amputee in Ecuador by two grad students with wonderful results. I’ve developed an open-source prosthetic hand and was a first author on a conference paper submitted to ICRA 2015. Aadeel and I have also explored the entrepreneurship aspect by participating in start-up programs and competitions such as Cozad, Onestart, and I-CORPS. I’ve had an awesome experience through this lab and will be continuing work with them. I received a complete research experience as I’ve been given both the rights and responsibilities of grad students. Below I will go into further detail on my work.
SCHOOL YEAR 2013-2014
As stated above I was fortunate enough to have co-authored a conference paper on sensory feedback for people with prosthetics, that was published in the Eurohaptics 2014 conference. In essence the device we developed gives linear skin stretch to indicate the position of a finger, or multiple, on a hand. This allows users to tell what grasp they are using as well as the aperture of the grasp without looking at the hand. Our design was extremely low-cost, energy efficient, and light weight. We were invited to give an oral presentation and live demo at the conference. The demo was scheduled for 2 hours but due to a huge interest the demo was run for 6 hours. Above you can find a video of me doing a demo of our work with a full explanation of our paper.
In high school I had made a 3D printed arm as an independent project. Upon arrival at UIUC I contacted Aadeel, a stellar grad student in the RMS lab, and because of the arm we were able to start this new line of research. Most of my work for this paper consisted of designing parts with CAD, building, and modifying this robotic arm we used as a test platform. I also did lots of Matlab programming to design our tests. Besides this I was a test subject and ran tests on the other subjects.
Most of the driving force for the summer research was due to the fact that Aadeel and I started a collaboration with a non-profit prosthetics group based in Ecuador. There is a huge need for low-cost prosthetics in countries like this and the vast majority of these are simple mechanical devices. Having worked with robotic arms we knew we could develop a low-cost myoelectric hand that could even be used with our linear skin stretch device as it was low-cost, energy efficient, and light weight. We based our design of the pre-existing open-source Dextrus hand and our end design cost was under $300 for a fully implementable system; this is comparable to the cost of a $30,000 – 100,000 commercial arm.
To get a fully functioning myoelectric hand that could be used on an amputee we needed to make massive modifications and new parts in CAD to get the mechanics of the hand to be effective and compatible with sockets. On the electronics side we had to design and prototype circuits to read and amplify the EMG signals used to control the hand, for motor control with logic, to add a user interface with buttons and an LED, interface with a micro-controller, and make a power circuit for both the logic and motors off one battery. On the programming side we wrote PD controllers for the motors, made a program to take EMG data and perform linear discriminant analysis to classify what grip the user wanted as well as train to define these classifications, write a task scheduler to simultaneously take EMG data, pick a grip, drive the motors, and repeat. Myself and Aadeel led a team of 3 undergrads to accomplish this in a little over a month and a half. I was heavily involved in every step and besides the hands on experience I gained knowledge in leading while meeting time-intensive deadlines, planning everything down, keeping good lines of communication to ensure no stops in development.
Through this project I developed skills and worked with programs such as rapid prototyping, 3D printing, several CAD programs, PCB design, SMT soldering, EAGLE, Blender, C/C++, and Matlab.
SCHOOL YEAR 2014-2015
I began more independent research on the Tact Hand, and open-source, low-cost, myoelectric prosthetic hand. Full details can be found on the Tact project page. In addition we are currently continuing our sensory substitution work and integrating it into the Tact hand.
Due to the success of our collaboration with ROMP, Aadeel and I wanted to explore commercial options in developing the prostheses. We were accepted into the NSF funded I-CORPS start-up program to develop a business model and do initial work to see if we had a viable start-up. We we’re successful in this program and made contacts with many giants in the prosthetics field to help direct us. Recently we were accepted as finalists in the prestigious OneStart bio-tech start-up competition. For this program we will be working with a business professional in the bio-tech world to help further develop our business model and ultimately pitch to venture capitalists. In addition we will be competing in the Cozad start-up competition at UIUC this spring.