This project aims to simulate and train the designed hardware-based Spiking Neural Network (SNN) to get the optimized
synaptic weights for efficient and effective performance. We set up the simulation environment necessary for training using Python and Brian2, a Python library specialized for SNNs. The model was used to simulate the dynamic behavior of neurons and synapses to understand their interactions and overall network behavior. We applied a training algorithm to optimize the synaptic weights and improve network performance.
Semester of Project:
Spring 2025
Team Photo:
Team Poster:
Problem Statement/Summary:
Fully simulate and train a hardware based Spiking Neural Network (SNN) using Python and optimize the synaptic weights on the neural network to be able to fine tune the performance of the SNN on a line-following EV car.
MS Teams Meeting Link:
https://teams.microsoft.com/l/meetup-join/19%3ameeting_ZTVmYzA0NDMtODY5NC00MTMyLTg3NjEtMjhhZDI2NGU1YWQ2%40thread.v2/0?context=%7b%22Tid%22%3a%224130bd39-7c53-419c-b1e5-8758d6d63f21%22%2c%22Oid%22%3a%22abb2769b-83e9-4396-a90e-d886c2995a87%22%7d
Project Department:
SOET
Project Presentation Video Embed Code:
Proprietary
Project Sponsor Website:
https://prf.org/
Project Sponsor:
prf
Team Contact:
inaba@purdue.edu