Robot Coverage Control by Neuromodulation

Kyle Harrington (1) , Emmanuel Awa (1) , Sylvain Cussat-Blanc (2) , Jordan Pollack (1)

1 Brandeis University
2 IRIT-VORTEX - Visual Objects from Reality To Expression

An important connection between evolution and learning was made over a century ago and is now termed as the Baldwin effect. Learning acts as a guide for an evolutionary search process. In this study reinforcement learning agents are trained to solve the robot coverage control problem. These agents are improved by evolving neuromodulatory gene regula- tory networks (GRN) that influence the learning and memory of agents. Agents trained by these neuromodulatory GRNs can consistently generalize better than agents trained with fixed parameter settings. This work introduces evolutionary GRN models into the context of neuromodulation and illustrates some of the benefits that stem from neuromodulatory GRNs.

Mots clés: Evolved neuromodulation; Gene regulatory networks; Baldwin effect

Domaines: Synthèse d'image et réalité virtuelle [cs.GR]; Traitement du signal et de l'image [eess.SP]; Traitement des images [eess.IV]; Vision par ordinateur et reconnaissance de formes [cs.CV]; Intelligence artificielle [cs.AI]

Kyle Harrington, Emmanuel Awa, Sylvain Cussat-Blanc, Jordan Pollack. Robot Coverage Control by Neuromodulation. International Joint Conference on Neural Networks - IJCNN 2013, Aug 2013, Dallas, United States. pp. 1-8. ⟨hal-01147304⟩