One of the seniors requested for me to collaborate with him on one of the components in his teams senior design competition last semester. The two teams at UNCA will be adhering to the following IEEE competition rule set and competing against each other. The winner of the teams would go on to compete in the actual IEEE competition.
In general, the teams need to build a bot that performs specified tasks such as entering a code in a keypad, pushing a button, grabbing rubber ducks, and much more. One of the seniors saw my research on the compliant gripper and my multi-material printing experience and asked for my help in adding a compliant gripper to their UAV. They wanted their UAV to fly and grab some of the rubber ducks, meaning we needed to make the mechanism as compact and as light as physically possible.
I was sent the following video to use as a template. I downloaded XYZAidan's model and imported it into fusion 360. I had to completely redesign the central hub to account for a different motor, as well as shrinking the entire model down. We tested many variations of this model, adding extra channels to increase articulation. This reminds me of kerf bending in woodworking, where slots are removed from one side of a board to be able to bend the entire piece to a desired angle. I even designed a small wedge to glue between these channels so we could test which had the best result without reprinting many times.
One of the biggest challenges I faced throughout this project was determining how to tension the arms equally. XYZAidan used bolts to wrap the thread around and tension that way. We ended up tying a nut to the end of the thread, and then looping that around the end many many times. This resulted in a friction fit during testing, and when we were happy with the results, we added some superglue to make it a permanent connection.
This video details an early version of their entire project. The video shows a 4 armed version of the gripper, very similar to in the video above. We eventually discovered that the ducks were pretty difficult to pick up in this way, and that maybe a "grabber" was better than a "gripper". We then removed two of the opposite arms, which greatly reduced the weight. I also then added these hook elements on the ends of the arms which would grab the underside of the ducks head, where there was an overhang it could rest on. I decided to print the main model in TPU, which produces the maximum elasticity and least plastic deformation from repeated use. I also decided to print the grabber pieces in PLA, which are more rigid and able to support the weight of the duck. The final piece came out to be just over 7 grams, which is a huge accomplishment for how well it works.