Numurus’ NEPI platform provides an open-source software foundation for developing smart systems, combining sensor integration, AI model deployment, automation, and data collection capabilities in a single edge AI environment.
Developing autonomous platforms often requires significant time spent building underlying infrastructure, including hardware drivers, data pipelines, AI frameworks, and control interfaces. NEPI addresses these challenges by providing these capabilities out of the box, allowing teams to focus on mission-specific applications.
Built on ROS and ROS 2, NEPI supports robotics, drones, subsea systems, industrial automation, and other edge AI applications where real-time processing and operation without cloud connectivity are required.
The platform includes hardware abstraction drivers, AI model management tools, event-driven automation, and a browser-based Resident User Interface (RUI) for system configuration and monitoring. A Docker-based NEPI Container also enables development and evaluation on standard Linux machines without dedicated edge hardware.
In a whitepaper, Numurus details how the NEPI platform is helping organizations streamline smart system development, from autonomous robotics and maritime operations to research and education applications.
The article also examines NEPI’s deployment with OceanAero’s TRITON autonomous underwater and surface vehicle (AUSV), where the platform supported AI-enabled maritime threat detection and inspection capabilities for the Defense Innovation Unit’s Unmanned Systems for Maritime Domain Awareness program.
