I have been working on developing the robot tentacle I demonstrated in class.
The first iteration was a bare proof of concept, and so the tentacle was not really mechanically connected end to end, except by the Paracord. It seemed that the next improvement would be to create a tentacle that is continuously connected (ie all one piece). This will hopefully aid in more stable and consistent control from the Paracord.
To that end, I designed a modular piece in Rhino (my first time using Rhino!)
This piece is designed with a male ball joint on one end and a female on the other, so the part can be duplicated and fit together in series to form a tentacle.
My plan is (was) to 3D print the original, and then cast/mold rubber to make the final, flexible part. (My first time 3D printing as well!)
This plan may work, but I might pause this to pursue another more simple solution.
Why not just use a pool noodle instead?
The disadvantages I foresee with this sectional design is that it is:
- Complicated–Molding this piece and then casting it in rubber materials is probably going to prove an involved, multi-step process.
- Heavy, if cast in rubber–This would limit the capability of motors to move the tentacle (they would have to do more work.
This approach may be suitable down the road, as the design becomes more detailed and iterated upon, however for now a pool noodle gives the following advantages:
- Simple–it’s already in one piece and ready to go!
- Very light–which gives maximum advantage to the controls.
The only modification needed for the pool noodle is to cut grooves for the control guides to sit in (those rings with holes cut in them that guide the Paracord). New control guides will also need to be designed.
I will be designing and fabricating this setup in the days to come, and will be demonstrating on Monday.
So in summation, I learned a lot from the Rhino/3D printing process, and will be saving its fruits for later, when the mechanical realities of the tentacle are more defined. The digital design/3D printing process is too specialized and overly ‘nice’ for this part of the project. I suspect the pool noodle approach will yield some very usable results.