The EMI Lab - TU Delft

The Efficient circuits & systems for Machine Intelligence (EMI) research lab focuses on designing energy-efficient digital AI hardware (ASIC & FPGA) for edge computing, emphasizing ultrahigh-speed communication, video/audio processing, robotics, and biomedical applications. 

We strongly believe that hardware-software co-design is the path to realizing the ultimate goal of intelligent edge solutions for complex real-world applications.

New master projects are available!

News

Dr. Chang Gao selected as an MIT Innovators Under 35

March 23, 2023

Dr. Chang Gao has been selected for MIT Technology Review's prestigious Innovators Under 35 Europe list. The annual list recognizes outstanding young innovators from across the continent who have made significant contributions to technology and its potential to transform the world. Complete list of IU35 Europe: https://emtecheurope.com/innovators-2023.

Dr. Chang Gao Receives the €203K MSCA-PF Grant

February 13, 2023

I am happy to announce that our Marie Curie Skłodowska Fellowship project, “AIRHAR: An Energy-Efficient AI Powered Portable Radar System for Human Activity Recognition," in collaboration with Prof. Francesco Fioranelli, Prof. Alexander Yarovoy and Prof. Leo de Vreede, has been granted.

We are thrilled to explore the possibilities of leveraging AI model compression and edge AI hardware in radar sensing. We are eager to engage in fruitful conversations with industry partners on potential collaboration opportunities in this area.

Dr. Chang Gao Awarded the MECA Prize

January 15, 2023

We congratulate Dr. Chang Gao, leader of the EMI Lab, for winning the 2022 Mahowald Early Career Award for his project titled "Accelerating Recurrent Neural Networks with Neuromorphic Principles." The project introduces efficient computing using the neuromorphic principles of spatial and temporal sparsity, leading to an energy-efficient accelerator for edge RNN computing. Gao's accomplishments have the potential to revolutionize applications in healthcare, wearables, and intelligent biomedical implants.

Research Highlights

Robotic Prosthesis Control

Collaboration with the AMBER Lab, Caltech

Spoken Digit Recogntion

Using the EdgeDRNN Accelerator

Neuromorphic Computing

Interfacing EdgeDRNN with a Silicon Cochlea

Recent Publications

Spartus: A 9.4 TOp/s FPGA-based LSTM Accelerator Exploiting Spatio-Temporal Sparsity (2022)

A 23μW Solar-Powered Keyword-Spotting ASIC with Ring-Oscillator-Based Time-Domain Feature Extraction (2022)

How to Find Us