Project

Back to overview

Brain-Inspired Chips for Edge Computing

Applicant

Nair Manu Vijayagopalan

Number

187863

Funding scheme

Bridge - Proof of Concept

Research institution

Institution of higher education

University of Zurich - ZH

Main discipline

Microelectronics. Optoelectronics

Start/End

01.11.2019 - 30.04.2021

Approved amount

159'558.00

Show all

All Disciplines (3)

Discipline

Microelectronics. Optoelectronics

Electrical Engineering

Information Technology

Keywords (6)

Neuromorphic; Chip design; Edge computing; Accelerators; CMOS; Artificial intelligence

Lay Summary (Italian)

Lead
Microcircuiti ad alta efficienza energetica per dispositivi dotati di piccole batterie
Lay summary
Durante il mio dottorato, ho lavorato sullo sviluppo di circuiti ispirati al cervello in grado di lavorare 500 volte meglio dei comuni processori, quali l’ARM Cortex-M4. Tramite simulazioni al computer, ho dimostrato che gli algoritmi implementati dai miei circuiti offrono una precisione molto maggiore dei predecessori in diverse applicazioni reali. Come risultato, la mia ricerca ha prodotto due brevetti e molteplici pubblicazioni peer-reviewed, oltre alla creazione di una startup, Synthara AG, nata con lo scopo di commercializzare i dispositivi da me progettati. L’obbiettivo di questa proposta è iniziare dei progetti-pilota con potenziali clienti operanti in diversi settori, quali veicoli a guida autonoma, smartwatches, controllo e gestione di allevamenti e fattorie, dispositivi e impianti medici, sicurezza personale e sul lavoro.

Direct link to Lay Summary

Last update: 07.06.2021

Lay Summary (English)

Lead
Energy-efficient AI chips for tiny devices with tiny batteries
Lay summary
During my PhD, I worked on the development of brain-inspired or neuromorphic architectures that performs more than 10x better than existing neuromorphic systems, especially when operating on sensor data streams such as audio data. I implemented these designs in two neuromorphic chips called ReASOn and Memoire. The goal for this project was to complete developing these technologies into a commercially-ready solution and initiate pilot projects with potential customers in a range of applications such as autonomous vehicles, medical implants, predictive maintenance, farm and livestock monitoring, and safety monitoring. I was very successful in achieving this goal over the course of the project. There were some technology challenges, because of some design bugs in the memoire chip. We identified workarounds to those issues and addressed them in a meaningful way. This allowed us to secure interested investors and customers for Synthara. We have completed technology transfer to Synthara and I am now working on the development of the next generation of this technology that is ready for large scale implementation. During my PhD, I worked on the development of brain-inspired or neuromorphic architectures that performs more than 10x better than existing neuromorphic systems, especially when operating on sensor data streams such as audio data. I implemented these designs in two neuromorphic chips called ReASOn and Memoire. The goal for this project was to complete developing these technologies into a commercially-ready solution and initiate pilot projects with potential customers in a range of applications such as autonomous vehicles, medical implants, predictive maintenance, farm and livestock monitoring, and safety monitoring. I was very successful in achieving this goal over the course of the project. There were some technology challenges, because of some design bugs in the memoire chip. We identified workarounds to those issues and addressed them in a meaningful way. This allowed us to secure interested investors and customers for Synthara. We have completed technology transfer to Synthara and I am now working on the development of the next generation of this technology that is ready for large scale implementation.

Lead

Energy-efficient AI chips for tiny devices with tiny batteries

Lay summary

During my PhD, I worked on the development of brain-inspired or neuromorphic architectures that performs more than 10x better than existing neuromorphic systems, especially when operating on sensor data streams such as audio data. I implemented these designs in two neuromorphic chips called ReASOn and Memoire.

The goal for this project was to complete developing these technologies into a commercially-ready solution and initiate pilot projects with potential customers in a range of applications such as autonomous vehicles, medical implants, predictive maintenance, farm and livestock monitoring, and safety monitoring. I was very successful in achieving this goal over the course of the project. There were some technology challenges, because of some design bugs in the memoire chip. We identified workarounds to those issues and addressed them in a meaningful way. This allowed us to secure interested investors and customers for Synthara. We have completed technology transfer to Synthara and I am now working on the development of the next generation of this technology that is ready for large scale implementation.

During my PhD, I worked on the development of brain-inspired or neuromorphic architectures that performs more than 10x better than existing neuromorphic systems, especially when operating on sensor data streams such as audio data. I implemented these designs in two neuromorphic chips called ReASOn and Memoire.

The goal for this project was to complete developing these technologies into a commercially-ready solution and initiate pilot projects with potential customers in a range of applications such as autonomous vehicles, medical implants, predictive maintenance, farm and livestock monitoring, and safety monitoring. I was very successful in achieving this goal over the course of the project. There were some technology challenges, because of some design bugs in the memoire chip. We identified workarounds to those issues and addressed them in a meaningful way. This allowed us to secure interested investors and customers for Synthara. We have completed technology transfer to Synthara and I am now working on the development of the next generation of this technology that is ready for large scale implementation.

Direct link to Lay Summary

Last update: 07.06.2021

Responsible applicant and co-applicants

Name	Institute

Nair Manu Vijayagopalan

Institut für Neuroinformatik Universität Zürich Irchel und ETH Zürich

Employees

Name	Institute

Nair Manu Vijayagopalan

Institut für Neuroinformatik Universität Zürich Irchel und ETH Zürich

Collaboration

Group / person	Country

	Types of collaboration

Institute of Neuroinformatics

Switzerland (Europe)

- in-depth/constructive exchanges on approaches, methods or results

Institute of Neuroinformatics - Prof. Giacomo Indiveri

Switzerland (Europe)

- in-depth/constructive exchanges on approaches, methods or results
- Research Infrastructure

Synthara technologies - Alessandro Aimar

Switzerland (Europe)

- in-depth/constructive exchanges on approaches, methods or results
- Industry/business/other use-inspired collaboration

Daedalean.ai

Switzerland (Europe)

- Industry/business/other use-inspired collaboration

Communication with the public

Communication	Title	Media	Place	Year

Media relations: radio, television

Press release for Seed funding of Synthara

International

2020

Talks/events/exhibitions

Synthara is part of ESA-BIC incubation

International

2020

Abstract

During my PhD, I worked on the development of brain-inspired or neuromorphic architectures that performs more than 10x better than existing neuromorphic systems, especially when operating on sensor data streams such as audio data. I implemented these designs in two neuromorphic chips called ReASOn and Memoire. The goal for this project was to complete developing these technologies into a commercially-ready solution and initiate pilot projects with potential customers in a range of applications such as autonomous vehicles, medical implants, predictive maintenance, farm and livestock monitoring, and safety monitoring. I was very successful in achieving this goal over the course of the project. There were some technology challenges, because of some design bugs in the memoire chip. We identified workarounds to those issues and addressed them in a meaningful way. This allowed us to secure interested investors and customers for Synthara. We have completed technology transfer to Synthara and I am now working on the development of the next generation of this technology that is ready for large scale implementation.