Learning experience maps on a neuromorphic chip with Khepera robot equipped with a dynamic vision sensor

In this project, we will build a neural network on a spiking neuromorphic chip ROLLS [1] that will learn a simple map of the environment. The chip will be interfaced to a small robotic vehicle Khepera-IV (k-team: https://www.k-team.com/mobile-robotics-products/khepera-iv) that will be additionally equipped with a neuromorphic “artificial retina” camera DVS (dynamic vision sensor: https://inilabs.com/products/dynamic-vision-sensors/). The project can deal with different aspects of robot control, neural learning on-chip, or event-based vision.

[1] Qiao, Ning, et al. "A reconfigurable on-line learning spiking neuromorphic processor comprising 256 neurons and 128K synapses." Frontiers in neuroscience 9 (2015).

Requirements

Programming experience in C++ or Python; interest in robotics, AI, and neuronal systems; electrical circuits understanding is not a must at this stage, but would be great for development of a tighter integrated neuromorphic system in the future.

Project level: Semester project, Bachelor or Master thesis.

Contact

yulia.sandamirskaya (at) ini.uzh.ch

© 2017 Institut für Neuroinformatik