Ok maybe we didn't have stagefright, but we were a bit nervous.
During this week we created and presented our entire 10 minute presentation for our Research Roundtable presentation. It was a great opportunity to practice showcase our research in front of an audience outside of our research group.
We divided up the sections based on our current experiences with the entire project. As a result I was required to talk about the push primitive and my summer work with the project. Each of us were required to talk for about 2 minutes on average and then as a group we fielded questions that were spawned as a result of our presentation.
Short Summary of what I contributed and presented on.
Since I was tasked with talking about the push primitive and summarizing my summer work I knew I would have to break down what a motion primitive was in much simpler terms. A great analogy that I try to use is that they are pieces to the whole puzzle and in this case the whole puzzle is the entire environment we would test the snake on. This way the audience can understand that each of the actions build on one another and become an entire motion plan for the environment. Second of all I had to summarize my summer reasearch, however, that requires a quick separate overview to get the gist of it.
Summer Research In A Nutshell
This past summer I got the opportunity to work in the Intelligent Vision and Automation Laboration as a full-time undergraduate research assistant. During this time I had to learn about the vision processing portion of the project, and also help include the current code into the entire process. As a result I was able to improve and finish the visual processing portion.
This caused me to learn even more about the vision portion of the project, since when the Kinect camera snaps the picture of both the snake and the environment it filters the snake body and the environement separately from the original captire. Also there was a massive reorganization of the scripts in the project folders. Taking an object-oriented approach I created two new classes to hold data that was just pertinent to those entities in the project. So for everything that is needed to model the Snake it will be held in the class SnakeModel, while everything that is needed to model the Environment will be held in the class EnvironmentModel. Each of these classes are used to either change the tailframe (otherwise known as the frame of reference attanched to the back end link of the snake robot) of the snake in real time by finding it in the picture or doing the computation to create the shape of the snake on top of the environment model and to make sure it is displayed relative to reality.
If you would like to check out the entire presentation click here.
Placeholder text by Nelson R.. Photographs by NASA on The Commons.