Jesse Lane

I used to keep a research journal here, but I’m moving all of my research posting to this here site. I figure that if I ever do anything interesting it’s more likely someone will find it here if they google me. Therefore, I’m going to start tagging research posts with the “research” category and personal posts with, you guessed it, the “personal” category. Now to go through all those previous posts and retag them. I’ll be posting an update soon about what I’m working on.

Every year the HCI department puts on a little shindig to show off the research, etc., that goes on there. It used to be the HCI forum and then it was up-sized to the Emerging Technologies Conference. At this conference the students are required to put up a poster on their research and stand by it to answer questions for an hour. Here is my poster for this year. Click on it to check it out in all of its glory.

ETC2009 Poster

I was kicking around to see if anyone has done any other simulation work with the Robotis Bioloid and I found a cool looking physics editor. Scythe appears to be a physics editor that is able to use a few existing geometry file formats and outputs file formats for some common physics engines. Here is the list of features. My favorite are the videos of ragdolls being catapulted into a brick wall. The video of a bunch of ragdolls falling down the stairs is pretty cool too, but I’m somewhat desensitized after Porrasturvat. Oh, and the best feature of all is that Scythe is being open-sourced (or will be by the time this is read).

Well, I can’t sleep. Thinking about robots and my dissertation. Kevin and I set up a mailing list with our buddies about our dissertations and John sent me an email with some great ideas and commentary. One of those ideas was to use a preexisting robot to make my life (and by extension those around me) easier. Since I’d like to do some research on autonomously developed robot body schemas I would like a robot that is reconfigurable. It should also have some decent sensing capabilities and be able to be either connected to a PC or run remotely by one. Enter the Robotis Bioloid. This baby looks pretty awesome. The expert system is about $3500 which is pretty pricey, although, not that pricey for a serious robotics project. It’s also possible to start with the comprehensive set which starts around $850.

I’ve also been thinking about pairing the Bioloid with a simulation environment. That way I could get to work before I even get a real robot. I’ve been comparing simulators/robotics platforms and right now the main contenders are Microsoft Robotics Studio and Player/Stage/Gazebo. They both have their pros and cons. Here’s a pretty good recent comparison of different robotics platforms.

I also need to keep in mind what the actual science is going to entail and get only what I need to run experiments. I need to avoid the temptation to get things because they are cool. Also, the more I have to do, the longer this is going to take and the crazier I’ll be at the end of it.

Ok, my eyes are finally starting to get a little bit heavy so I’m going to head off to bed.

So I started reading this book Embodiment and Cognitive Science by Raymond W. Gibbs Jr. (Amazon link here) which was suggested to me by Alex Stoytchev. I haven’t gotten very far yet (p. 26, in fact) but it seems to be pretty good so far. He’s currently discussing what people notice about their bodies.

When I run uphill, I certainly feel the muscle tension in my legs, and the expansion of my lungs as I struggle to breathe. But I have little awareness of the hair on my head, the movement of my hands, the sensations in the pit of my stomach. Thus, the body does not appear to consciousness as a normal object of awareness as we actively engage with our surroundings.

Maybe Gibbs covers this later in the book, but if we take his attempt at reconciling the body with the mind and the environment seriously (which I do) then it seems to me that the body does appear as a normal object of awareness. The reason is that when we percieve what has been classically considered “the environment” we do not attend to the flood of incoming stimuli. We focus the fovea of our eye at one thing at a time. We typically attend to one voice at a time in a crowded, noisy room. Similarly, we attend to a single, or few bodily sensations at any one time.

I am very interested in body schema and body image, especially as it applies to robots, so creating an attentional system that either encompasses or integrates multi-modal attention or determines separate “modules” for each sensing modality may be an interesting avenue for future research. I would also like to extend the contingency work that Kevin and I did for Narcissus to systems that use many more sensors in multiple modalities.