The startup develops AI-in-sensor processing technology that enables direct coupling of sensors to convolutional neural networks, significantly reducing latency, power consumption, and costs in edge computing applications. This technology provides ultra-low power and ultra-low latency performance, enhancing the efficiency of AIoT devices compared to traditional solutions like memristors and resistive RAM.
Funding
$30.2M raised to dateRaised to date based on public sources. This may differ from the amount the company actually raised and is based only on what is publicly available on the internet.
Founders
Product
Problem
Traditional edge computing solutions often suffer from high latency and power consumption due to the need for digitizing sensor data and transferring it to a separate processor for AI inference. This separation increases system complexity, cost, and time to deployment, hindering real-time AI processing at the edge.
Solution
AIStorm offers AI-in-Sensor technology that directly couples sensors to convolutional neural networks, eliminating the need for analog-to-digital conversion and external processing. By processing data within the sensor itself using charge domain processing, AIStorm significantly reduces latency, power consumption, and system costs. This approach enables real-time, always-on AI processing at the edge, facilitating applications such as high-speed imaging, audio keyword spotting, biometric extraction, and vibration sensing. AIStorm's integrated circuits (ICs) and wireless modular solutions cater to diverse applications, including building automation, retail analytics, and security systems. The company also provides services for dataset development, labeling, and modeling to support the deployment of AI-in-Sensor solutions.
Target Audience
AIStorm's primary customers include OEMs and system integrators in the consumer electronics, automotive, security, medical imaging, retail, and industrial automation sectors, seeking to implement low-power, real-time AI processing at the edge.
Features
- Charge domain processing for ultra-low power and ultra-low latency AI inference
- Direct sensor coupling, eliminating the need for analog-to-digital converters
- AI-in-Sensor Imager ICs with integrated imager, AI, AEC, power management, LED driver, Bluetooth, WIFI, and ARM processor
- High-speed imaging capabilities, up to 260,000 frames per second, for facial recognition and machine vision
- Audio KWS & Sound Spotter ICs for keyword and phrase spotting, as well as sound event detection (gunshot, glass break, baby crying)
- Biometric Extraction AFE ICs for heart, EEG, EOG, and muscle ID in wearables, automotive, and textiles
- AI-in-Sensor Vibration Solution ICs for industrial, infrastructure, and machine reliability monitoring
- Wireless modular solutions for building automation, retail, and security, including face recognition imagers and always-on sentry imagers
- CodeZero GUI for simplified AI development and optimization on resource-limited edge devices