The potential of artificial intelligence (AI) and machine learning (ML) in embedded systems is rapidly growing, leading to the development of innovative application solutions and products. However, creating AI-based systems can present challenges.
STMicroelectronics has recently introduced NanoEdge AI Studio, a cutting-edge ML technology that helps accelerate innovation by providing an optimal ML library for an AI project with minimal data. This technology is demonstrated in a video that showcases how it collects current and vibrometer information to train an ML model for detecting anomalies in a system.
NanoEdge AI Studio is a PC-based pushbutton development studio that is compatible with Windows, Linux, and Ubuntu. It eliminates the need for advanced data science skills by generating four types of optimized tinyML libraries: anomaly detection, outlier detection, classification, and regression. These libraries can be combined and linked together to create a comprehensive edge AI solution. Anomaly or outlier detection can identify equipment issues, classification can determine the problem source, and regression can offer valuable insights. Inputs can include various signals like vibration, pressure, sound, magnetic field, and time of flight, which can be combined from multiple sensors within a single library or across multiple libraries.
The microcontroller handles learning and inference using the NanoEdge AI on-device learning library, making edge AI processes more streamlined and reducing development efforts significantly. Interested individuals can explore the related links for more information on this revolutionary technology.
