Contact Information:
Mark Rubianes - Team Leader mprubian@ncsu.edu
Benjamin Davis - Meeting Scribe bjdavis9@ncsu.edu
Ben Kozel - Sponsor Contact blkozel@ncsu.edu
Ayush Luthra - Purchasing Contact aluthra@ncsu.edu
Jason Cuddeback - Website Manager jwcuddeb@ncsu.edu
Sponsor Meeting Time:
Fridays 12:30 PM
Group Meeting Time:
Tuesdays and Thursdays 10:00 AM
Assignments:
Sponsor: NewCareTec
Mentor(s):
Jim Meckstroth jmeckstroth@gmail.com
Dennis Ewald
Zaher Mashal
Background of Company:
NewCareTec is an IoT start-up with a mission to make residential living facilities smarter to benefit the safety and well-being of elder residents and augment the care-giver staff coverage.
Project Background and Need:
NewCareTec is an IoT start-up founded by technologists with a mission to make residential living facilities smarter to benefit the safety and well-being of elder residents and to augment the care-giver staff coverage.
According to the CDC, every second of every day, an older adult (age 65+) suffers a fall in the U.S.—making falls the leading cause of injury and injury death in this age group. One out of four older adults will fall each year in the United States, making falls a public health concern, particularly among the aging population.
Staffing around the clock is a challenge for the growing population of residential living facilities.
Sensors and communication technologies associated with deploying IoT sensors can enhance safety by making a facility “smarter” without requiring additional technology “interaction” by the elder residents and without risks to privacy or hacking.
Problem Description/requirements (List of high-level functionality you expect to see in the end product):
Prototype a battery powered sensor node to characterize distress audio in normally unoccupied spaces (e.g. stairwells) for potential safety benefit. The device is to avoid any image capture and minimize any other privacy intrusion or hacking risk.
Key Requirements for the sensor node:
+ Sensor node using ST Micro STM32WL55 SOC sub-GHz MCU. (e.g. Nucleo-64
STM32WL55JC17 Arduino compatible development board).
+ Add on Microphone to capture Audio
+ Add on PIR sensor to detect human presence (and activate active listening for sounds
of distress) to extend sensor node battery life.
+ “Edge Impulse” embedded Machine Learning (ML) inference engine to characterize
sound and detect “abnormal sounds”, including spoken “help”.
+ LoRaWan communication of distress “alerts” to Raspberry Pi/RAK wireless LoRaWan
Gateway with network server and application server running Node-Red
+ Training of “Edge Impulse” ML inference engine at a number of spaces within doored
areas (e.g. Stairwells) to demonstrate improvement of ML characterization with a goal
of >80% accuracy.
+ Gateway Node-Red Dashboard to log events and send notifications
+ [Optional] Prototype circuit board and/or packaging of sensor node (including MCU SOC
development board, Audio Capture Mic, PIR sensor, battery, Antenna)
+ [Optional] Gateway cellular communication of alerts via Blues Wireless “notecard” to
“Notehub” with routing to a cloud based “Dashboard” (e.g. Tago IO or similar)
Required Skills / Desired Competencies (in order of importance):
Understanding or desire to learn implementation of Edge Impulse (or similar) machine learning on microprocessors, as well as Node-Red rules engine
Configuring wireless nodes using LoRaWan protocol
Embedded programming and hardware
Machine learning
Wireless communications
Programming experience
Analog circuit design
Networking
PCB design and layout
How often and how would you be available to meet/chat with the team (Zoom, Hangout, Skype, Phone)?
At least bi-weekly. In person and/or Zoom.
Will the students be required to sign any of the following before they start with the project? NDA / CDA / IP assignment / IP protection form
IP assignment agreement will need to be signed.