Model for a sensor inspired by electric fish

Frédéric Boyer (1) , Pol-Bernard Gossiaux (2) , Brahim Jawad (1) , Vincent Lebastard (1) , Mathieu Porez (1)

1 IRCCyN - Institut de Recherche en Communications et en Cybernétique de Nantes
2 IN2P3 - Institut National de Physique Nucléaire et de Physique des Particules du CNRS

This article reports the first results from a programme of work aimed at developing a swimming robot equipped with electric sense. After having presented the principles of a bio- inspired electric sensor, now working, we will build the models for electrolocation of objects that are suited to this kind of sensor. The produced models are in a compact analytical form in order to be tractable on the onboard computers of the future robot. These models are tested by comparing them with numerical simulations based on the boundary elements method. The results demonstrate the feasibility of the approach and its compatibility with online objects electrolocation, another parallel programme of ours.

Mots clés: Electrolocation; active perception; bio-inspired sensor; model reduction; electrokinetics; elec- trostatics; resistance matrix

Domaines: Automatique / Robotique

Frédéric Boyer, Pol-Bernard Gossiaux, Brahim Jawad, Vincent Lebastard, Mathieu Porez. Model for a sensor inspired by electric fish. 2011. ⟨hal-00614081v2⟩