Development of neuromorphic technologies for Visual Inertial Odometry

Scientific-Disciplinary Group

09/IINF-05 - Information Processing Systems

Description

The activity will focus on: i) becoming familiar with neuromorphic continual learning algorithms like Continual Learning Prototypes (CLP) in Python; ii) re-implementing CLP-SNN in Python and evaluate it on perception tasks; iii) implementing real-time continual learning mechanisms for SNNs on a PULP-based microcontroller architecture (GAP9), such as the one used in the Bitcraze Crazyflie microdrone. The student will identify the most suitable neuromorphic continual learning strategy and port it to the target GAP9 platform. The proposed evaluation benchmarks will target robotics perception tasks, including optical flow estimation and visual–inertial odometry (VIO).Activities will be developed within the ROBOMIX2 project.

Job posting website

https://bandi.unibo.it/ricerca/incarichi-di-ricerca

Funding body

ALMA MATER STUDIORUM - UNIVERSITA' DI BOLOGNA - - CENTRO RICERCA SISTEMI ELETTRONICI INGEGN.INF. E TELECOM."ERCOLE DE CASTRO"

How to apply

Other

Selection process