24 March 2010

Post date: Oct 27, 2010 5:24:13 PM

Here we conducted trials with 4 robots, organised as 2 pairs: epucks 9 and 12 are initialised with a movement pattern which describes a square trajectory and epucks 29 and 32 initialised with a closed 4-segment L- shaped pattern. We have constrained the movement-imitation algorithm (both in estimating the turn angles, and in enacting movements) to multiples of 90 degrees. The effect of this constraint is evident in the qualitatively high fidelity copies of both square and L-shaped movement patterns.

Here is a plot (created by loading the tracking data into Excel), from run 2.

What is interesting is that robots in the 9/12 pair have, from time to time observed and stored the meme patterns from the 29/32 pair, and vice-versa. Because of the weighted selection criterion the odds are against one of these being selected and enacted but, nevertheless, we see a square pattern enacted by epuck29 (lower left) and an L-shaped pattern enacted by epuck9 (upper right).