James Coleman is one of those people that you see everywhere. He’s constantly moving, constantly on time, and constantly happy. He’s always there to help you when you are utterly and completely confused and he’s probably TA-ed a class that you’ve taken or taught you some life-changing earth-moving new skill. If it hasn’t happened to you yet, it will. And when it does, you can quickly and awkwardly look up this cheat sheet of recent James Coleman endeavors! So that you can say…”Hey, James, thank you for being awesome. I hear you like drawing robot legs.”
I’m in this studio with Jose Selgas. We are doing bio-inspired plant walls. The end result will be an 8 x 8 x 8 cube. There will be 5 panels and the different groups will be focusing on different elements of plant biology. My team’s focus is on a natural composite that is formed for single serving edible plants. The idea is that you can go to the store and purchase this pressure-formed packet (of seeds etc…) that is curated by type of meal (for instance, a salad) for growing and consumption. Basically it’s a single serving meal packet. The way it works is that there is a series of small green houses that act as modules that you can attach to various places. The cool part is that we are playing with coconut husks and exploring how we can form coconut husks with natural matrices. We are going to try to use a cornstarch glue that we are developing to be the matrix that hold the fibers together. The coconut husk is useful in that the fibers can be laid omnidirectionally. Depending on how strong the matrix is and how strong the fibers are, it can become fairly structural.
The idea is that little bulbs can be grown in a natural formed pot. Instead of having a separate pot, the pot is the grow material. As long as we format it correctly you can plug these packets in and rotate them based on the sun. The way you interact with them is that you can read the description of what they are on the back and spin them around to have them facing out or in!
I think one of the benefits of going to MIT is that you get to invest a lot of different things but that is never fully explored so…I would like to talk now about robots.
I’m taking Robotics with Professor Harry Asada and I’m learning all of the things about robots that aren’t fun. Robots do a lot of things for you because they have been designed by companies. For example, if you want a robot to go to a certain endpoint, you can type in a position and it will go there. Right now, I’m learning about all of the math that goes into robotic movement. I’m trying to create the framework for a robot to take the endpoints and actually move there. The thing is, I’m finding out that it takes pages and pages of math to get to where you want it to go. In this class we are making robotic legs. The idea is that there are exoskeletons that clip onto a pair of legs (we are using half scale 3d printed legs). There are four motors and the four motors are basically the legs themselves. We are writing software and creating the mechanism to make them walk. The practical application would be that someone who is wheelchair bound can put these exoskeletons on and walk. What’s really cool about it is that I get to make these really awesome drawings of funny little robot legs and tons of calculations. It’s really fun stuff.
I’m TA-ing “How to Make Almost Anything” and Sheila’s Networked Sensors class. Sheila’s class is about how you can use low power sensors to gain information about environments. We are looking at what she calls subnatures (basically things that are overlooked unless you have data about them). It’s kind of like rendering the invisible visible. It’s a fun class and she’s partnered with Joe Paradiso, who runs the Responsive Environments group in the Media Lab. Sheila is bringing an interesting question to the table about what to sense. Her argument is that we need to think about why we are sensing and less about what we are sensing. Subnatures is one of their required readings and it is the last thing that I’ve read so that I can be a more active part of the ongoing class discussions.
Other than that I’m reading a lot of technical journals about robots because I’m going to Russia next week for a machine building workshop in Moscow. I took Neil Gershenfeld’s “How to Make Something that Makes Almost Anything” class and I got involved with an awesome group. An organization called MISI (Moscow Institute of Science Innovation) has invited us over for the workshop. I’m giving a talk on the first day and on other four days we are working in a lab to create a more flexible set of tools to make other tools. It will be a kit of parts that you can join together so that machines can make any number of other machines. Instead of one-offs, they will be component parts that can be used to build a variety of other tools!