Embedded systems are increasingly incorporating artificial intelligence (AI) and machine learning (ML) technology, leading to the development of innovative products and solutions. However, creating AI-based systems poses challenges for developers. NanoEdge AI Studio is a ML technology offered by STMicroelectronics that aims to simplify the process of developing AI solutions. The platform generates an optimal ML library for a project with minimal data requirements, eliminating the need for advanced data science skills.
In a video demonstration, NanoEdge AI Studio collects current and vibrometer information to train a ML model for detecting anomalies in a system. This technology operates on a PC-based pushbutton development studio that is compatible with Windows, Linux, and Ubuntu. Developers can easily create four types of optimized tinyML libraries: anomaly detection, outlier detection, classification, and regression.
By combining and chaining these libraries, developers can create comprehensive edge AI solutions. Anomaly and outlier detection can identify issues in equipment, classification can pinpoint the source of the problem, and regression can provide insightful extrapolations. The platform supports various inputs such as vibration, pressure, sound, magnetic field, and time of flight, enabling the integration of data from multiple sensors into a single library or across multiple libraries simultaneously.
The microcontroller conducts both learning and inference processes using the NanoEdge AI on-device learning library. This reduces development efforts significantly by simplifying the edge AI workflow. This technology holds great promise for accelerating innovation in embedded systems and facilitating the creation of advanced AI-based products.
In conclusion, NanoEdge AI Studio is an excellent tool for developers looking to streamline their development process when creating AI-based systems for embedded systems. With its simple interface and powerful functionality
