Prosthetic Arm

Having some spare time within my college schedule, I decided to pickup a long term personal project that I could eventually use as a senior research and design project. Something that I have always been intrigued by is the field of artificial limbs and humanoid robotics. Trying to explore projects within this field, I chose to concentrate on a robotic arm as it is challenging, but manageable given my experience.

​

However, I didn't want to just simply mimic what has already been done several times over. I wanted my arm to distinguish itself from the plethora of designs out there. Researching various designs, a pattern that I found was that not a single arm chose to mimic fingers with a rotary joint. All the fingers were fixed to linear motion. Therefore, these hands had a limited degree of movements and could not articulate completely like a human hand when it came to finger intensive tasks such as typing or playing the piano.

 

So, I planned to create a robotic arm that allows for the lateral motion of fingers. Rather than embarking on the full hand at first, I started off by designing a proof of concept that illustrated the concept of rotational finger movement. 

​

I designed a prototype rotary hinge using 3D printed parts, fishline and servo motors. Modeling my design after the human hand, the nylon fishline served as tendons while the servo motors mimicked the function of muscles.

​

Using this hardware and an Arduino program, I was able to demo the full range of motions a human finger is capable of. The finger could not only extend and retract, but also rotate to any position.

​

With the success of this initial prototype, I moved on to working on the full scale iteration with all 5 fingers. 

​

While I initially used strings to control left and right movements of fingers, I found it very difficult to incorporate this mechanism space wise when including all 5 fingers. Thus, I engineered a new mechanism that uses a spur gear system to shift fingers left and right. This geared method is much more space efficient and easier to control as well.

​

Another problem that I encountered during the first iteration testing was the difficulty in tensioning the nylon strings. They would either be too loose or too tight. To prevent these issues on the full scale, I designed custom string tensioner pulleys. Two screws on top of each pulley serve as the method of tensioning. Once a string is wrapped around, these screws can be tightened/loosened to increase/decrease tension.

​

My prototype uses servo motors and a custom spur gear to move fingers left and right. The nylon strings connected to servo motors fold and unfold the fingers. Through these degrees of freedom, several hand motions and positions can be replicated.

 

I plan on taking this prototype further by figuring out a way to control it using signals sent from the upper arm or the brain. Stay tuned for future updates and in the meantime check out the video below!

​

​