Posts by author:

August – September 2011

by admin on September 21, 2011 · 95 Comments

I resume my work on thesis in the end of august by implementing advanced logging into my C++ projects. Then, I wrote a python script so I have infrastructure to collect data from experiments and store them in my “experiments library“. I made several experiments with various parameters of CACLA algorithm and Neural networks architecture [...]

June – July 2011

by admin on July 20, 2011 · 92 Comments

After I finished all my exams and took a rest for a couple of weeks I returned to my master thesis. So what did I do in this period: added separate neural networks that acts as forward and inverse model, which both learns from data gathered during CACLA learning (to reach object). created a C++ [...]

Hand terminology & Grasping model

by admin on April 17, 2011 · 101 Comments

I found useful pictures of hand movement terminology. These terms are essential to understand iCub’s hand and fingers DoF. I have read about FARS model of control of reaching and grasping (Fagg, Arbib et al.) Another article which copes with grasping models is Models for the control of grasping (Oztop and Kawato). Then I’ve drawn [...]

iCub touch and pressure sensors

by admin on March 29, 2011 · 105 Comments

iCub has 6 touch sensors on each hand (5 are on the last part of fingers and 1 is on the palm). In the iCub Simulator, output from these sensors is available at port /robotName/touch and it is streamed as a list of 12 values Example of output: [lpam] 0.0 [rpam] 0.0 [lind] 0.0 [lmid] [...]

February – March, 2011

by admin on March 8, 2011 · 95 Comments

At the beginning of February I implemented neuralnetworks and reinforcement learning C++ libraries which I use in my iCub application. The most of the time I spent by implementation of the application which uses the CACLA algorithm to train iCub to reach objects. I’m also planning to create some UML package and class diagrams to [...]