With less than two weeks left of my stay in Bloomington, I’m now hurrying to get everything that I planned for in. These days my time has been mostly occupied by:
[a] Setting up meetings with a number of people to talk about my work, potential future work, their work, etc: I have met with Linda Smith, Hillel Chiel, and William Timberlake. From each of them I have obtained useful feedback. I will be meeting this week with Luis Rocha, Peter Todd and John Kruschke.
[b] In collaboration with Bryan Daniels, Randy and I have been working intensely on ideas relating to CTRNN parameter space in relation to the evolutionary techniques. Basically, we have been thinking about putting Randy’s recent theoretical work to practical use in evolutionary robotics. Ideas involve: different forms of seedings for different tasks, how to choose the ranges of parameters for evolutionary search, mutation operators that orient towards the regions of richest dynamics, etc.
[c] Testing various robotics kits for the upcoming undergraduate robotics course. I’ve focused on the vEx kit: built the robot and got a CTRNN to control it online. Now I’m working on getting it to run from Macs – even if it means running Windows on it :(.
[d] Last but not least, I have working on the writing of the associative learning work. This is work in collaboration with Inman and Randy. The writing itself has helped me develop a clearer ‘story’. At the moment, I’ve focused on the results section, and it is about 75% there. All of the figures are now there. I have to move on to the rest of the parts before I can send a copy around for feedback.
I’m glad to say that the visit has turned out to be even more fruitful than expected.