The latest report from ABI Research looks at how embedded chipset vendors are increasingly prioritising Neural Processing Units (NPUs) for IoT applications due to their efficient handling of neural network workloads.
As implementers seek greater insights and intelligence at the edge, NPUs are expected to capture a larger share of overall shipments, surpassing traditional Microcontrollers (MCUs). ABI Research projects that chipset revenues from AI-dedicated silicon for IoT applications will exceed US$7.3 billion by 2030.
“NPUs for TinyML applications in Personal and Work Devices (PWDs) are already well established. However, they are still nascent outside of this device vertical, and major vendors are only just introducing this type of ASIC to their embedded portfolios,” said Paul Schell, Industry Analyst at ABI Research. “By screening PWDs, we provided greater insight into our modelling for IoT applications, which spans 15 verticals, including the most significant, namely Smart Home and Manufacturing.”
On the software front, comprehensive MLOps toolchains have become essential for vendors, including start-ups like Syntiant, GreenWaves, Aspinity, and Innatera. The investment in software offerings often parallels hardware R&D, as seen with Eta Compute’s partnership with NXP to license their Aptos software platform. These innovations democratise TinyML deployment by reducing the need for in-house data science expertise.
Incorporating highly performant architectures like NPUs and certain FPGAs into embedded devices will broaden the range of applications able to run on-device, from object detection to simple object classification for machine vision use cases, as well as some natural language processing (NLP) for audio-based analytics.
“Along with the trend in larger edge form factors such as PCs and gateways, this will contribute to AI’s scalability by reducing networking costs and the reliance on cloud. As such, we expect the TinyML market to grow as it capitalises on these innovations, spurred largely by major industrial sites upgrading their IoT deployments, the growing intelligence of vehicles, and smart home devices,” added Schell.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by commenting below or visiting our LinkedIn page.