There are significant advantages to running machine learning on the sensor, including cost and power savings.
MOUNTAIN VIEW, Calif. (PRWEB) May 18, 2021
Qeexo, developer of an automated machine learning (ML) platform that accelerates the development of tinyML models for the edge, today announced that it is working with STMicroelectronics to enable developers to create machine learning models for ST’s Machine Learning Core (MLC) sensors, so that inferences can run right on the sensor, without the need for a microcontroller. This feature will be available to users in Q2.
Traditionally, due to limited computation power, memory size, and battery life, building machine learning solutions for edge devices had been challenging. Qeexo AutoML solves this. Its one-click, fully automated platform allows customers to rapidly build machine learning solutions for edge devices using sensor data. By moving machine learning to embedded processors and now sensors on edge devices, developers can improve privacy, latency, and availability.
“Qeexo continues to demonstrate technical leadership in the embedded machine learning space by automating machine learning on tiny, resource-constrained devices – this time, on a Machine Learning Core sensor, independent from an MCU,” said Sang Won Lee, CEO of Qeexo. “For use cases that can benefit from machine learning, but do not have access to MCUs due to cost, power, latency, or infrastructure constraints, there are significant advantages to running machine learning on the sensor, including cost and power savings.”
“Many IoT solutions developers are looking to easily add embedded machine learning to their very low power applications and need help to bridge the gap from concept to prototype to production,” said Simone Ferri, MEMS Sensors Division Director, STMicroelectronics. “We put MLC in our sensors to reduce system data transfer volumes and offload network processing. Qeexo AutoML can help unlock the benefits of inherently low-power sensor design, advanced AI event detection, wake-up logic, and real-time Edge computing.”
Qeexo also announced that it is launching a model converter that can take machine learning models in the ONNX format to optimize them for embedded devices. For customers who already have a machine learning team, and who have worked on and have existing machine learning models, they can use the Qeexo Model Converter to make them smaller and more optimized for embedded devices. The technology will also make it easier for developers who want to compare the performance of their hand-built models against the models automatically created with Qeexo AutoML.
In addition, Qeexo is now offering a machine learning consulting service to help clients jump-start their projects. Qeexo will first work with clients to develop and deploy commercial-ready machine learning solutions, then tailor Qeexo AutoML to fit customer needs. Qeexo will provide the knowledge transfer necessary for client teams to continue to use Qeexo AutoML for current and future projects.
Qeexo is the first company to automate end-to-end machine learning for embedded edge devices (Cortex M0-M4 class). Our one-click, fully-automated Qeexo AutoML platform allows customers to leverage sensor data to rapidly build machine learning solutions for highly constrained environments with applications in industrial, IoT, wearables, automotive, mobile, and more. Over 300 million devices worldwide are equipped with AI built on Qeexo AutoML. Delivering high performance, solutions built with Qeexo AutoML are optimized to have ultra-low latency, ultra-low power consumption, and an incredibly small memory footprint. For more information, go to http://www.qeexo.com.