Buy Now

QuickLogic Blog

QuickLogic and SensiML Enable AI at the Endpoint for the Masses

You might have seen the exciting announcement we made this morning regarding immediate availability of our new Merced EOS™ S3AI Hardware Development Kit (HDK), which enables quick and easy development of endpoint Artificial Intelligence (AI) and Machine Learning (ML) applications. This new kit includes everything an endpoint application developer might need, including QuickLogic’s EOS S3AI SoC, an accelerometer, gyroscope, and magnetometer, along with two digital (PDM) microphones on an Arduino form-factor compatible board – making it easy for users to add different sensors or other 3rd party modules.

The Merced HDK links seamlessly with the SensiML Data Analytics Toolkit (free 90-day evaluation period included). This toolkit uses integrated machine learning technology which allows it to support a complete design flow from sensor data collection through automatic feature extraction and classifier development. Specifically, it efficiently identifies which complex feature extraction algorithms are ideal for the customer-labeled data sets and generates the AI models and classifiers needed for code development. The AI model that is subsequently generated is uniquely optimized and partitioned to fit in the EOS S3AI hardware, including the embedded FPGA, Flexible Fusion Engine, and MCU blocks. The tight integration of hardware and software ensures that the AI algorithms created with the toolkit are optimized for the EOS S3AI to ensure efficient and low power AI processing.

We are very pleased to say that this kit has been a huge success in the marketplace. Even though we just introduced it, we’ve already have orders for over seventy HDKs to various countries around the world. In fact, we have a strong backlog of orders and have been actively working with our channel partners and distributors to address the pent-up demand.

It’s clear that there is a strong global desire to add AI to endpoint applications, and that our Merced HDK is a great vehicle for accomplishing that goal. It provides a very easy and low cost approach for OEMs who want to evaluate the benefits of AI and quickly develop a unique solution for their endpoint AI and machine learning applications, without needing an army of data scientists or AI experts. Examples of such applications include predictive maintenance, structural health monitoring, and manufacturing process control, but there are many more as nearly any IoT endpoint can benefit from added local intelligence.

However, endpoint applications are notoriously difficult to implement due to power and bandwidth constraints. The combination of QuickLogic’s EOS S3AI embedded compute capability and low power consumption with the SensiML’s Data Analytics Toolkit complete development flow and our Merced HDK provides a complete and low power solution available to large and small customers all over the globe. To learn more about our HDK, visit https://www.quicklogic.com/products/quickai/. More information about the SensiML Data Analytics Toolkit can be found here https://sensiml.com/wp-content/uploads/SensiML-Technical-Overview-Brief.pdf.

Leave a Reply

Your email address will not be published. Required fields are marked *

button - scroll to top