An important goal of learning is to improve the efficiency of an organisms interactions with its environment, thereby increasing its chances of survival. The Neuromorphics Lab aims at integrating recent advances in neuromorphic engineering, computational modeling, and robotics, to design robotic agents capable of interacting in a natural environment and in real time. We are designing whole brain systems models in the MoNETA project and gradually porting them in mobile robotic platforms. Projects in robotics include the design of compact, low power neuromorphic hardware for visually guided navigation and obstacle avoidance, and interfacing adaptive robots and humans via EEG.
Figure MoNETA implemented in a mobile robot navigates toward a salient target.
|input to robot visual system||optic flow|
Figure Optic flow is a powerful cue to help a robot navigate, avoid obstacles, and estimate its distance from salient objects in its environment.
Below is The Neuromorphics Lab VIsually GUided Adaptive Robot (VIGUAR), a step towards the implementation of complex memristive-based brain in mobile robotics platform.To achieve this ambitios goal, we are initially developing models that run on single GPUs, two GPUs, and GPU clusters.