Cupertino, CA, June 22, 2022 – Aizip, Inc., a technology leader in artificial intelligence (AI) for applications on the Internet of Things (IoT), the so-called tiny machine learning (TinyML), announced today the general availability of Aizip Intelligent Module (AIM) for sensor fusion applications.
Driven by the fast progress of AI technology, sensor fusion has received increasing attention in recent years. Sensor fusion is the process of combining disparate sources of sensor data to reduce prediction uncertainty so as to enable more complex tasks. One example is to improve location-prediction accuracy by combining GPS signals from satellites and street views from cameras. The resulting information has less uncertainty than would be possible when these sources are used separately.
While sensor fusion has been used in areas such as self-driving vehicles, it hasn’t yet found its way to the IoT market. The very limited memory and processing power in the low-cost semiconductor IoT devices make deploying sensor fusion a daunting task. Renesas Electronics, a leader in IoT semiconductor devices and intelligent solutions, and Aizip formed a joint team to bring sensor fusion onto IoT platforms. Leveraging their deep market understanding, extensive TinyML expertise, and automated design tools, the team has developed technologies that simultaneously offer the high performance and processing efficiency required by advanced IoT applications.
In a demonstration of their sensor fusion capabilities, the team developed a Person Access System (PAS) leveraging the RA6M3 microcontroller device from the widely scalable 32-bit RA MCU family that uses an Arm Cortex-M4 core. Aizip Intelligent Module (AIM) integrates various AI functionalities such as person detection, face detection, face recognition, and speaker identification. The combined accuracy of person recognition is significantly higher than that achieved by face recognition or speaker identification alone. Based on a flexible architecture, the functions in PAS can be provided as a whole or in a variety of combinations to meet various customer needs.
“Aizip has showcased their superior TinyML technologies in this amazing demo,” commented Kaushal Vora, Senior Director, Business Acceleration & Ecosystem of Renesas. “While it is already challenging to put one TinyML model in an IoT device, the team actually put four models in one device.” The approach allows significant cost reduction and power saving over existing solutions that use much more powerful yet expensive semiconductor devices. Renesas has a strong focus on artificial intelligence of things (AIoT) and has worked closely with Aizip as part of their TinyML ecosystem.
“We’re very excited to see the PAS project going well,” said Yan Sun, CEO of Aizip, Inc. “Renesas is a worldwide leader in MCU and IoT devices. We’re honored to partner with them in the fast-growing market of TinyML.” Renesas’s high-performance MCUs have been widely recognized to be a leading choice for the IoT market. The excellent development environment makes the design and integration quick and efficient, according to Aizip.
Renesas debuted PAS at Embedded World 2022 in Nuremberg, Germany. The successful demo of sensor fusion in a compact MCU device represents a major milestone in pushing more intelligence into IoT devices. With the mission to make the world intelligent and sustainable, Renesas and Aizip are working with their customers in bringing the TinyML technology to a wide range of commercial applications.
With a focus on sensor fusion, Aizip’s new AIM product series will consist of two or more models from its portfolios of Aizip Intelligent Vision (AIV), Aizip Intelligent Audio (AIA), and Aizip Intelligent Time-Series (AIT). Aizip provides TinyML models to its IC partners and module and system customers worldwide. Aizip is committed to bringing low-cost and high-performance intelligence for the benefit of society worldwide. For more information, please contact info@aizip.ai.
* * *
About Aizip, Inc.
Aizip, Inc., develops AI models for edge applications. Based in Silicon Valley, Aizip provides models with superior performance, quick turnaround time, and excellent ROI. These models can be used in a wide range of applications for an intelligent, automated, and connected world. For more information, visit www.aizip.ai.