Growing Aspinity’s total funding to more than $19 million, the round also includes new strategic investor and partner Unitrontech, a distributor of automotive semiconductors and key partner of leading automotive manufacturers.
Founded in 2015, Aspinity has taken a new approach to power-conscious AI used in devices that continuously analyse real-time sensor data. Its RAMP (Reconfigurable Analog Module Processor) technology platform merges the ultra-low power benefits of analogue processing with the sophistication of machine learning and the versatility of software programmability.
The company’s first product, the AML100 analogue machine learning core, enables always-on sensing solutions at near-zero power that accurately detect and inference from raw, analogue sensor data. The AML100 keeps the MCU and other digital processors asleep until meaningful data are detected, focusing device power exclusively on information that truly matters.
The software-programmable chip can address many different inferencing applications with Aspinity’s python-based software development kit that allows user-friendly development, testing, and verification of analogue machine learning models.
Aspinity and Unitrontech aim to deliver high performance, near-zero power after-market and integrated AI solutions for automotive applications. For example, an always-on dashcam solution that leverages sensor fusion and AI to detect and record only the events that are important to the car owner (i.e., ignores sounds or vibrations external to and not related to the car). In this application, the AML100 can be programmed to detect impacts and intrusions in general, or to detect and alert to specific disturbances such as glass break.
“Our strategic partnership with Aspinity reinforces our commitment to providing the most innovative solutions to our automotive customers,” said Richard NamKoong, CEO, Unitrontech. “Automotive surveillance devices, such as dashcams, are gaining momentum in the automotive market, and we need to be able to sense, collect, and process data in unattended vehicles without draining the battery. The combination of Aspinity’s analog machine learning technology with our automotive solution expertise provides our customers with market-ready intelligent systems.”
“Aspinity’s technology enables the lowest power AI solutions,” said Tom Doyle, CEO and co-founder, Aspinity, Inc. “No other AI chip can interface directly with analogue sensors and inference while the data are still in their native analogue domain, keeping downstream computing resources focused solely on relevant data. This is the key to minimising power consumption and to a future of ubiquitous AI systems.
“We are working with customers around the world to integrate this capability into home security systems, automobiles, and condition monitoring equipment. This investment allows us to collaborate with strong partners like Unitrontech and further develop our technology for new markets.”
