Return to Independent Study
02.08.17 Independent Study, Journal 4
02.08.13 Wednesday, Meeting in the Coaction Lab
Meeting with Gene in the Coaction lab at 1:30. I did not get there until 2:00 because I was lost and walking all over campus.
The meeting was productive, Gene is good at discussing ideas but he is also good at getting us back on-track when it goes off too far. I appreciate both things a lot. I think in a lot of ways, that is what I need most from independent study. I need a little direction and I need to keep the idea generation in check. I don’t want to end up with a thousand ideas and no implementation.
I feel a lot better about this project now. I feel like I can use this semester and summer to make it something that could be a really great large scale project for next year.
That also works out well with the origin project. If I can get the first part of that project done this semester, then the remaining aspects of it could be side projects when I work on robotics. The remaining projects are far less technical. They are things I already have the skills to do. They are also less expensive. I might need some funding but probably not a lot. I could devote possible grant funds to the robots (which I really do love more).
Look into Festo, a Robotics Research a German company that does a lot of work on life-like robotics, bionics
There is a free version of MaxMSP called PureData. It doesn’t have all the bells and whistles and the interface is not as smooth but it is open source. Gene says that the original builders disagreed about whether the program should be open source or not. The result was two different versions.
We worked out a semester plan. This was very useful.
My first assignment is to use Processing to set up digital version of the ecosystem to show how the robots would behave. Use this time to learn about digital worlds. Make two organisms. One is an emitter and one is a receiver. See how they interact,
Spend two weeks building the first robot prototype. Build 2 robots. One robot should be the sender and the other an emitter.
Consider a 2 wheel and caster or 3-wheel design.
3-wheel design: One wheel in front with a servo motor. Then two wheels in the back with one motor so that one is a slave wheel.
Gene drew a picture on the whiteboard and talked through the build. Simply seeing someone do that made the process more accessible. He said we could meet next week to discuss parts to buy after I do the processing assignment.
Refine the previous build. Explore behaviors.
Build the first robotic system with 4 robots.
Revise the 4-robot system. Refine the overall architecture, smaller footprint, build/design custom PCB
Custom PSBs could be a delaying issue. I do not know how long it takes to get them. I hope that it does not prevent my robot progress. I feel like I need to fully complete all of this by the end of the semester. I hope that some of the process goes faster than expected so I can order PCBs a little sooner.
I have added this step. I have an extra week that we did not account for because of Spring Break. I would like to spend that week working out the robotic environment. I need to decide if the robots require their environment or are simply more “at home” in it. I prefer option 2.
I will also use this time to create viewer questionnaires and/or surveys so that when I do the first test run I can gather data about how the robots are being perceived by viewers.
I will also figure out how to design and who will do the viewer testing.
The viewer testing and data compilation.
I will form a presentation book to document the process, experiment and results. This may be imprint or online. That decision will be made based on the quality of the data.
I will eventually create a whole book of this project. I don’t want to print and bind it if I feel I don’t have a good handle on the format. However, I also want to do this as I go rather than at the end of what should be a very long project. Doing it all at once would be too overwhelming. I will attempt to create a flexible enough format that I can somehow print and assemble it as I go.
Meeting Notes on Arduino Parts
Screw Shields (buy at least 2 sets)
Mini Breadboards that can allow the robots to be adaptable
I think my first phase design should consider adaptable robots from the start. I believe that even if I have to make very large robots, this would allow the next phases to go more smoothly. I can shrink the robots later!
I also think I should get around to making that Arduino parts inventory so I can reference it when I am designing and/or meeting with Gene.
Could I somehow make an LED cube or Cube-like object a part of the robotics system? I did consider building them an interactive system. I also have a somewhat amusing idea of creating an abstract god-concept for the robots. Somehow, I think an LED cube could very much do that. What if the robot environment contained some kind of central figure made of light? It probably wouldn’t be a cube, but something made with the same kind of system. Maybe the robots could respond to things that the cube does. The cube could even be a trickster. It could misdirect them and confuse the system. Maybe it could even be influenced by viewers, allowing the viewer to become “gods” to the robots. It’s a very sad concept to me because I believe that even if the robots psychological responses appeared happy, the viewer would know then that it was false. For instance, what if the viewer removed robots from the system but another viewer had god tell the remaining robots that they were still there. The robots who were removed would attempt to reunite with their system but they would not be able to do so. They would show fear responses while their social group ignored them. It would all be very complex coding, but maybe someday I could be good enough… who knows. On a simpler level, it would be interesting to simply use the lights as a way of monitoring and interacting with the robots. That would not be so hard.
02.11.17 Project Proposal Revision & Processing Tutorials
I revised will revise my robotics project proposal to include more details and a better time line.
Then I listened to a series of videos on Processing and practiced working with the software.
I went through a series of 15 Processing tutorials. I have 5 more to go. After that, I estimate that it will take me apx 10 more hours to learn what I need to learn and 5 hours to write the code for my simulation.
This is Part 1 in a 20 video series of Processing tutorials. They are 3-8 minutes each. They are easy and well explained. The series takes you step by step through enough practice to be able to understand how to work with the language. From there, more complex tutorials will be needed to meet the specific project.
I am trying to stay on-task with processing, but I am also concerned that I will need to order robot parts. I know I can do the processing assignment. I am less certain about the robot time line. Since one assignment does not necessarily build on the other I am going to work on both at the same time and hope I still end up with the processing simulation complete by Wed.
This is pretty awesome. This is kind of similar to what I had in mind for the caterpillar robots I thought I might eventually put in my robot ecosystem. It isn’t on my current project list, but it’s cool that I found it.
This robot is made of rectangular segments. I was thinking of something with round segments and ball and socket joints. I thought it would be cool if they could have “brain” units that controlled everything but “body” units that could come apart and rejoin with other robots. It’s way too hard for me now, but I like the idea.
Here is another one by MIT made from cubes
It doesn’t use servos. It has an inertial actuator with a spinning mass moving inside it. When it moves, the spinning mass stops and the motion is applied to the robot. The inertial actuator also allows jumping! Commands go through a radio. They want to put the algorithms on the modules themselves.
This video was useful because they show the process of designing it. I like that. I find the logic useful.
Then YouTube took me to the MIT Self-Actuation Lab and it started showing me foldable materials that are pretty much what you do when you make origami tessellations. But now I am off topic way too far. Back to robot designs.
This is neat because they give you all the files and instructions to build it. That might be a good learning activity and then the finished robot could be altered for prototyping.
Similar to what I need but I want to do it myself.
Robot Parts Needed
Place to Mount Sensors – Under, Sides, Back, Front, Top
Maybe go ahead and include speakers since they are big and need space to fit.
Large Solderless Breadboard
PRODUCT ID: 443
Breadboard: 175mm x 145mm / 6.9″ x 5.7″
Metal Plate: 221mm x 183mm / 8.7″ x 7.2″
Height (with Breadboard & Metal Plate): 10mm / 0.4″
Binding Posts Spacing: 19mm / 0.75″
Wow, I am really scattered today. I just remembered the department grant is due Friday so I need to do that too!
I am just researching to support and possibly develop my argument for the project.
The Man Who Would be God, “Giving Robots Life”
I read this before but lost it so here it is again
02.13.17 More Robot Plan, Grant & Processing
I have way too much to do! I made a list but everything on it has the same basic importance. Damnit. Ok. If that is the case, I’ll just pick from the list and focus as long as I can. Then I’ll switch.
I worked on the IMFA grant last night. Then I E mailed Owen and Gene about it. Owen write back and said: “Destin – whole grant application should not be more than a page plus your budget. Really, we want a description of the project and your need for support and then how does this help you develop as an artist and what is the significance of the project (this could be to you in your development, or to others) – really what we are after is why we should fund this (who benefits and how do they/you benefit).”
Obviously, I need to rewrite. I was approaching it like the GSG Grant. On the bright side, I came up with some awesome thoughts I can use later.
I rewrote the abstract of the MFA Grant. It is short and sweet but I think I can make it shorter and sweeter. I need a better idea of budget first.
Back to robot design so I can budget
I found a good robot base design on Instructables: Plywood Robot Multipurpose Base