This is the engineering summary of an open reference architecture. The paper, its object model as JSON and the model register are free to reuse under CC BY 4.0: https://codeatoms.ai/vegetation-mapping-lidar-us/. The operator is described by class, never by name.
An agriculture and earth observation operator in the United States needs a vegetation baseline for its site. Aerial surveys usually arrive as a folder of tiles and point clouds. Next season, someone rediscovers what was flown, at what accuracy, and under which contract.
Here is how one flight becomes a baseline the next season can compare against.
1. Fix the acquisition, then everything else
One manned flight within a week either side of 1 July. 3 inch orthoimagery in four bands, red, green, blue and near infrared, plus USGS Quality Level 1 LiDAR at a minimum of 8 pulses per square meter. The leaf on window and the pulse density are what make the vegetation analysis possible at all.
2. Make the flight mission the focal object
Sixteen typed objects, from the site and its units, watercourses and dams, through the camera, the LiDAR sensor and the GNSS and IMU chain, to the orthoimagery, the classified point cloud, the elevation models and the QA/QC accuracy report. The contract, the invoices and both project managers are objects too. Every product traces back to the mission, the sensor and the accuracy report behind it.
3. Stay inside the operator's GIS
Four named sources enter through two adapter families into the operator's ArcGIS Enterprise. Nothing runs in a cloud the operator does not control.
4. Serve no model
Vegetation analysis stays with the operator's own analysts. The model register is empty and no compute is bought. The survey firm prices the acquisition against the operator's own task table.
5. Accept against evidence
The operator's project manager accepts each deliverable against the QA/QC accuracy report, not against a delivery date. The next flight compares against a versioned baseline.
Full design, figures and the object model: the paper. DOI: https://doi.org/10.5281/zenodo.23186673
Designed on CodeNinja Praxis, the CodeNinja Atoms platform for designing physical AI systems. The object model imports into Hyper Ontology, which turns it into a living system. Load it yourself with the open hyper-ontology loader.
Top comments (0)