Neural Decision Processors for Smart Home Applications

Syntiant Engineering, 2021–22

Liaison(s): David Garrett, William McDonald ’20
Advisor(s): Daniel Contreras
Students(s): Marco Conati(TL-S), Alexandra Loumidis(TL-F), Yoo-Jin Hwang, Kevin Kong, Kip Macsai-Goren

Syntiant Corporation produces ultra-low power, high-performance neural decision processors for use in machine learning applications. This project aims to integrate a 9-axis inertial measurement unit(IMU) with one of their Neural Decision Processors, the NDP120, in order to create a event detection system capable of monitoring door motion in the home for security purposes. To this end, the team is developing a custom PCB along with ML-enabled software for estimating door states and identifying key events.