This is a summary of one of the sessions that took place as part of the MIT Connected Things WFH Edition. May, 7th 2020.
Interviewer:
Mark Thirman, Board Chair, MIT Connected Things
Featuring:
Michael Ellenbogen, founder of Evolv.
What Evolv Does?
They offer walk through devices that automatically detects firearms or threats. The device is an example of an IoT device that uses AI / Machine Learning.
What makes the device an IoT device:
– Smart sensors
– Server software that analyze the data from the sensors
– 4G LTE modem for connectivity
How does the system gets smarter over time using Machine Learning?
– The more systems are deploy
– The more data are collected
– Some of that data is pushed to the cloud to help training the algorithms
– They test the new algorithms extensivly
– The new algorithms are pushed back out to the field.
What are the main use cases?
– Stadiums, schools, venues and concert halls: where in a short period of time before the show starts thousands of people needs to be scanned
– Workplace violence
– Casinos
What’s a possible Covid-19 Use case?
The same system could be used to detect people with fever or to limit the number of people entering a place per certain time period.
What impact does the 5G has on IoT products?
A lot of the processing is currently taking place on-prem (on premise) on the device’s CPU/GPU’s, with 5G, more processing and more data analytics will be done on the cloud without having to scale up the physical device.
More information on Evolv: https://www.evolvtechnology.com/
More information on the MIT Connected Things WFH Edition: https://www.mitconnectedthings.com/
Trademark Legal Notice. All product names, trademarks and registered trademarks are property of their respective owners.
