After deciding on proceeding with the slider design and doing the necessary calculations, we printed out the first prototypes for the index finger of our hand.
As the original design had an extra degree of freedom, this gave rise to issues such as the middle piece not sitting properly after a few flexions and extensions, as it is able to move freely. Thus sometimes, we will observe the middle slider piece sliding above the first finger tip part. Thus, we further improved on the design in order to tackle this problem. We tried to do this by adding a stopper in the middle piece of our slider design. In our first attempt, we added the stopper nearer to the piece for the finger tip. However, we found that this made the motion of the finger quite unnatural as the proximal interphalangeal joints (PIP) of the finger would have to extend first before the distal interphalangeal joint (DIP) can extend (as can be seen in Figure 2 below). Naturally, both the DIP joint and PIP joint are gradually extending at the same time.
Figure 1: The unnatural movement of the parts in the first design where the stopper was placed near the end, as shown from frame 1 to 3.
In order to solve this problem, decided to shift the stopper in the middle slider piece to the middle of the entire piece instead. This will still help to reduce the degrees of freedom, and prevent the middle slider piece from sliding above the first finger tip part. At the same time, this will not restrict the motion on the DIP.
Figure 2: In the top is the first draft of the middle piece. Here, the stopper sits near the end of the piece, preventing the other part attached at that end from translating. In the bottom is the subsequent draft of our middle piece.
Figure 3: Second attempt at modifying the middle slider piece
The performance of the modified slider piece proved to be satisfactory, and the movement can be seen in video below. This allows the finger to move in a more natural manner than what is shown in Figure 1.
Some Issues Found
We also tried to control the finger part using a servo motor. However, as we were using a PE braided line, many kinks were formed after several rounds of testing. There will be a need to further address this issue perhaps by putting the wire within Teflon tubing. If that does not work, we might need to change the wire that we use to control the finger parts.
Figure 4: Kinks in the wire when testing the prototype with the servo motor