structuring a summer project: what am I gonna do :( [updates]
Okay so here's the thing. Most of our summer plans have been cancelled due to corona virus so I think now's a good time to learn the things I've wanted to learn.
See I know where my strengths right now are, and my strengths right now are at (in engineering) the system-level. I'm very good at looking at the system from a top-down view but I want to get better at things like the design of an individual component, the math behind the design and etc. So that leaves me with two things to do. (1) I want to review 2.001 notes and look at 2.002 materials on MIT OCW, and (2) I really want to give myself a design project that I can complete so I can feel confident in my design abilities.
Because the other problem is, I'm used to designing things with paper and pencil, and then just building them, not running them through Solidworks or MATLAB simulations to figure out how well they would work or how to optimize the design. So I want to get a lot better at this.
I've also been working on chip, and while of course I'm going to continue working on chip, working on it (and through 2.12 both) has made me realize many different ways I'd re-design the system if I was given a chance to. So I thought, why not give myself this chance this summer. To redesign it properly.
So here's my plan. After finals:
(1) start with the basics of what we need chip to be able to do/design requirements/what's the POINT of chip because we don't want to just make another MIT Cheetah or Spot. (I'm thinking it's going to be to make these kind of robots really cheap).
(2) Design this thing on paper, do some technical drawing or something which is the method I'm used to.
(3) Put the design in Soildworks and start doing some math. So we'll do some hand-calcs on the system as well as FEA because I need to get better at FEA. And we'll re-design the parts as needed.
(4) Come up with a manufacturing plan, spec the costs and etc. While we're doing all this we're going to design the electrical system (which shouldn't be too hard as I have experience with this, but we'll do it properly and we'll show the power calculations).
(5) Take this Solidworks assembly and port it over to MATLAB and we can do some control design. Maybe some code generation. Modeling torque control and etc.
(6) Finally write up everything I did over the summer into one document just so I can compile my work and then we'll have a design for chip3!
So that sounds like a lot but I also have the whole summer and I'm not starting from scratch. I'm choosing to design Chip3 not because I'm going to build a Chip3 right now but because I know enough about the Chip system where I think I'd be able to learn a lot about design itself without having to learn what the system is supposed to do first.
And I've already got some ideas going for the new actuator/etc! I know what I want to fix after building the first chip so why not get to it?
I'm also going to take this post to talk a little about why I want to make a chip3 and what I think a platform like that could be good for.
I firmly believe that robots can do so much good in our world and especially with the versatility of legged robots. Here's the thing right. Legged robots can be used in almost any environment because of their ability to handle terrain. So what I really want to do is think about how we can produce inexpensive legged robots that can help in developing nations to either (a) find sources of water, or (b) for things like sustainable agriculture. The reason for this is legged robots have been used in developed nations to increase farm productivity and if we can make an inexpensive system where one legged robot can increase the productivity of a whole village's food system, or better, one that can farm where we couldn't originally put farms by extracting nutrients from other locations, that could make a large difference right?
I tried using robots for good in high school and it went well but it wasn't the most sustainable idea and it didn't go far enough. Now I think I'm better equipped to try a project that can do some real good, and I'm not saying I'm going to succeed, I think let's start this idea and see where it takes us. Inexpensive, legged farming robots, for developing nations. This could be project chip in for good!
Anyways, the first things we need to do are get chip two walking so we can think about farming with chip two and where that can help, and then start working on the design for chip three.
But what this means is we need to make chip three SUPER cheap, robust, and functional. So that's the goal of that design.