Name in original language
-
Initiative overview
The DeepGreen project addresses the need for AI systems that can run directly on small, low-power devices. Traditional AI often relies on cloud computing or large data centers, which are not suitable for devices like sensors, wearables, or industrial equipment that require fast, local processing. This creates a barrier to using AI in many real-world applications where speed, privacy, or limited connectivity are concerns.
In response, DeepGreen partners launched the Eclipse AIDGE platform in 2023 — the first open-source software platform dedicated to embedded AI. It allows developers to design, train, and deploy AI models directly on hardware devices, without sending data to external servers. The platform supports energy-efficient AI and helps reduce latency while improving data security. Being open source, it is freely accessible and can be adapted by developers, researchers, and companies of all sizes.
The platform was developed through collaboration between research institutions, technology companies, and open-source communities. It includes tools for building AI models, testing their performance, and running them on different types of hardware. This helps make embedded AI easier to develop and more reliable in everyday use.
By connecting expertise from various fields, the DeepGreen project has created a practical tool for applying AI outside the cloud directly where data is generated and decisions need to be made.
