--- title: Physical AI system design, port drone watch emoji: ⚓ colorFrom: blue colorTo: yellow sdk: static pinned: false license: cc-by-4.0 short_description: Physical AI system design, port drone watch tags: - physical-ai - sovereign-ai - germany - maritime-and-ports - drone-detection - ontologies - reference-architecture --- # System design for physical AI: Port Drone Watch An open reference architecture for physical AI in maritime and ports in Germany: a docked drone network at two port sites turned into detections, dispatched work items with a named responder and a living map, fifteen objects over five source systems, five open-weight models on three tiers of the authority's own hardware, and people deciding every response. By CodeNinja Atoms. The full paper renders on this Space. The operator is described by class, never by name. PDF and source files: https://github.com/muhammadumar89/codeninja-research. Ontology and register: https://huggingface.co/datasets/CodeNinjatools/drone-detection-dispatch-port-germany-ontology. Canonical page: https://codeatoms.ai/drone-detection-dispatch-port-germany/ DOI: https://doi.org/10.5281/zenodo.23635729. CC BY 4.0. Made with [CodeNinja Praxis](https://codeatoms.ai/praxis/), the CodeNinja Atoms platform for designing physical AI systems, and [Hyper Ontology](https://codeatoms.ai/hyper-ontology/), which turns the object model into a living system. One of the CodeNinja Atoms [sovereign AI reference architectures](https://codeatoms.ai/sovereign-ai-reference-architectures/): every published design with its sector, country, compute and three-year cost. Its object model sits with the other [industrial ontology examples](https://codeatoms.ai/industrial-ontology-examples/).