Models and robots:

Apart from using classical experimental techniques, such as electrophysiology, for unraveling the neural mechanisms involved in polarized skylight detection, we have also employed the synthetic approach. To study how insect POL-neurons view the natural sky, we have constructed an opto-electronic model of a POL1-neuron. To test ideas how e-vector orientation could be extracted from the signals of three POL1-neurons, we have observed the orientation behavior of mobile robots equipped with polarization compasses. A neural network proposes how these signals could be processed by actual neural circuits rather than by mathematical algorithms as used in the robots.

Opto-electronic model of a POL-neuron
Mobile robots with polarization compass
Neural network