Announcing our next tinyML Talks webcast! Siddharth Tallur from IIT Bombay
will present Edge-compatible machine learning algorithms for vibration condition based monitoring of machines on July 13, 2021 at 5:30 AM Pacific Time.
Condition based monitoring (CBM) leverages sensor measurements for measuring state of health of an asset and autonomous diagnosis of faults to trigger remedial actions. Complex deep-learning models pose memory and processing speed constraints for edge implementation, while cloud-based computing poses high latency, high cost of data transmission and storage and privacy threats. In this talk, I will present recent work from our group on a light-weight, non-redundant, edge (Raspberry Pi) implementable CNN algorithm that utilizes vibration sensor measurements for fault event estimation of machines. The model was trained and tested on two publicly available and widely studied vibration datasets for rolling element bearing faults, with near perfect accuracy for both binary as well as multi-class fault classification. I will also discuss a light-weight CNN autoencoder implemented on a PYNQ Z2 FPGA board for anomaly detection in an unsupervised learning framework, with less than 10,000 trainable parameters and approximately 90% accuracy.
Dr. Siddharth Tallur is an expert in physical and chemical sensor systems, MEMS and photonics, and embedded systems. His Ph.D. thesis research on low phase noise RF opto-mechanical and opto-acoustic oscillators won the best thesis award in the Electrical and Computer Engineering Department at Cornell University in September 2013. Following the completion of his PhD in 2013, he worked at Analog Devices Inc. in Wilmington, MA, USA, where he conceived and led the characterization of novel gyroscope designs and mixed-signal circuit architectures for inertial motion-sensing applications. He joined IIT Bombay in Department of Electrical Engineering in November 2016, where he currently serves as Associate Professor and faculty-in-charge at the Wadhwani Electronics Lab (WEL).
==========================
IMPORTANT: Please register here
Once registered, you will receive a link and dial in information to Zoom teleconference by email, that you can also add to your calendar.
We encourage you to register earlier since on-line broadcast capacity may be limited.
Note: tinyML Talks slides and videos will be available here on the tinyML Forum and the tinyML YouTube Channel afterwards for those who missed the live session. Please take a moment and subscribe to the YouTube channel today
Feel free to ask your questions on this thread and keep the conversation going!