Final Project Tracker

Please take a look here, to see the final system.

--------------------------------------------------------------------------

System: Minimal deployable machine

I have been quite interested in tape-springs (see links below). I think it could be interesting to use them to build a very light-weight deployable machine (e.g. no extrusions, just three 'end-blocks', each with an extendable (tape) actuator, meeting in the middle to support a tool of some sort). Ideally, the individual modules might be robotic/self-assembling ...

Compliant/deployable actuators:
tape-springs (link, link, link, paper, Pellegrino lab).

Questions:

(a) anchoring, blocks need to anchor themselves ~ e.g. screw themselves into the ground

(b) stiffness of the extendable actuators and system ~ what kind of processes would this allow (and at what resolution)? e.g. piling things, pick-and-place, others? Also, what kind of application is interesting, in the context of deployability/unstructured environments.

(d) do I want to focus on making a good actuator for the tape-springs for now?

Component 1: actuator design

Some things to consider: how to put pressure/release pressure on the tape when it is actuated vs being driven by other actuators, e.g. do we need 1 motor + 1 solenoid? Or choose a motor based on good brake/holding torque?

Reference, Vineet mentioned Jeff Hoffman's group's project on deployable towers for the moon, also here. This in the reference list: link.

Thoughts for encoders/motion feedback: - Optical flow-sensors (e.g. like in an optical mouse?) - Could I hack an optical mouse? (link); (link); link -> googled this chip Github; and got to this link (already 12 years ago!); also more recent projects that have used this sensor though, link
Questions for programming: - .gcode for abritrary (uneven) delta set-up?

Position feedback:

Exploring the optical sensor via old computer mice harvesting ... I took apart a bunch of mice. The two nicest (= simplest / easiest to disassemble) ones I found also seemed to work off sensors that looked the closest to the Arduino-examples I had found online (the MCS-12085 sensor). The chip inscriptions were different however: "OM50ND" and "LJFD3180C". I tried to reverse engineer/sketch out the circuit from the Inland mouuse I had taken apart, and it looked like the pins for the 0M50ND mapped out similarly to the mcs-12085.

Adding a swivel base:

The torsional stiffness of these tapes will not be great, so I think I will need good bearings to allow the actuators to swivel easily/follow the path of the tool-head.

Test: custom deployable tape

Redesign: spool for custom folded tape

New design prototype

Changing track: zippers!

Unsatisfied by the torsional stiffness of the tape prototypes and returning to some experiments into velcro-like surfaces during the fall class of HTMAA, I decided to explore zipper-inspired mechanisms as an alternative. While less known than other linear actuators, there are plenty of precedents for chain and zipper-like actuators with impressive applications (e.g. to raise theater stages/other platforms - zipper-chain, single-chain version, plastic version). However, I have not seen examples of these kind of actuators being used to ceeate compact personal fabrication/desktop machines.

---