Nexus Observatory Sign inCreate account
Nexus Observatory

/

Edge compute & AI

/

Scientific compute

Run the model where the data lands.

Scientific compute, inside Edge compute & AI: EO foundation models, training data and labels, land cover and change detection, hazard and exposure mapping and more. Every item is traced to its specification, from OGC Training Data Markup Language for AI (TrainingDML-AI) to CEOS Analysis Ready Data framework, and runs on the open Observatory Protocol.

Plan with usArchitectureThe protocol
Technologies
9
At the frontier
3
Protocol functions
3 of 8
Industries served
8

Reference design, drawn to scale · drag to turn · hover a part

01What it is

Foundation models for Earth observation and geospatial.

AI from model to accountable operation: edge inference and model optimisation, foundation models for Earth observation and geospatial work, time-series and sensor AI, multimodal fusion and supervised agentic operations, federated and privacy-preserving learning, physical AI and robotics within stated red lines, MLOps, evaluation and model risk, and the governance frameworks that apply to each. Compute placed where data is produced and decisions are needed: the device, far, near and regional edge tiers, multi-access edge computing, modular edge data centres including the AI tower in a box, servers and accelerators, storage, power and cooling, ruggedisation, fleet orchestration, OTA and GitOps, and the hardware trust that makes remote sites safe to run. Each technology below carries what it is, what we do with it, and the standard it is drawn against.

AI systems

EO foundation models

Large pretrained models over optical, SAR and multi-temporal imagery, fine-tuned for many downstream tasks.

We design, integrate and operate.

OGC Training Data Markup Language for AI (TrainingDML-AI) · CEOS Analysis Ready Data framework

AI systems

Training data and labels

Documented, licensed training sets with known label quality and geographic coverage.

We design and operate.

OGC TrainingDML-AI · ISO 19157-1 (data quality)

AI systems

Land cover and change detection

Classification and change maps with accuracy reported against reference data.

We design, integrate and operate.

ISO 19144-2 (Land Cover Meta Language) · ISO 19157-1

AI systems

Hazard and exposure mapping

Flood, fire, landslide and subsidence mapping that feeds risk and response decisions.

We design, integrate and operate.

CEOS Analysis Ready Data framework · ISO 19157-1

AI systems

SAR interferometry and deformation analysis

Measuring millimetre ground and structure movement from radar phase, with learned noise and unwrapping aids.

We design and integrate.

CEOS Analysis Ready Data: Normalised Radar Backscatter

Frontier · AI systems

Geospatial embeddings and similarity search

Vector representations of places and scenes that make large archives searchable by example.

We design and integrate.

OGC GeoAI Domain Working Group

Frontier · AI systems

Super-resolution and gap filling

Generative methods that sharpen or fill imagery, labelled as model output with stated uncertainty.

We design.

ISO 19157-1

Frontier · AI systems

AI weather and climate emulators

Learned models that emulate numerical weather prediction at much lower compute cost.

We integrate and operate.

WMO Integrated Processing and Prediction System (WIPPS)

AI systems

Geospatial model metadata

Machine-readable model descriptions with inputs, bands, resolution and intended use.

We design and operate.

STAC Machine Learning Model extension · ISO 19115-1

02Architecture

Where it sits, and what it speaks.

Scientific compute draws on 2 domains and 16 layers of the spectrum; the highlighted layers carry this area. It speaks 3 of the protocol’s 8 core functions.

Edge tiers and multi-access edge computing

9

Modular edge data centres and the AI tower in a box

8

Servers and accelerators

8

Storage and edge data management

6

Cooling and ruggedisation

10

Kubernetes and fleet orchestration

8

OTA updates, GitOps and software supply chain

8

Confidential computing and hardware trust

8

Edge inference, model serving and optimisation

10

Foundation models for Earth observation and geospatial

9

Time-series and sensor AI

8

Multimodal fusion and agentic operations

7

Federated and privacy-preserving learning

7

Physical AI and robotics (within red lines)

7

MLOps

7

Evaluation, model risk and AI governance

10

01JoinIdentity, attestation quote, signed policy.

02AdmitAdmitted for a stated scope, nothing wider.

03DescribeWhat, where, how and by whom, with calibration.

04TimeOne clock across the grid, with its error bound.

05ObserveEvery value with quality and provenance.

06QualifyTested against the specification, state on record.

07FederateSigned exchange across grids and trust domains.

08OrchestrateSoftware and configuration by signed plan.

03On the grid

Every site joins one observing grid.

The open networks on the grid today, over real relief and the geodesic cells we count them into, filtered to the space domain this area serves. A site built here enrols into the same grid.

04In the field

Where this runs, sector by sector.

4 deployment scenarios from the sector research that rely on scientific compute.

Government and public safety

Wildfire camera network and fire-weather stations

Camera towers with edge inference and co-located fire-weather stations, utility- or agency-funded.
Industrial

Utility private LTE for the grid edge

Distribution automation, distributed-energy control and wildfire-risk weather on a dedicated band.
Agriculture

Farm or cooperative agromet network with crop models

Compact stations and soil probes feeding irrigation, disease and frost models.
Energy

Utility grid-edge private LTE

See industrial.

05Open and enterprise

Open by default. Enterprise by contract.

The open counterpart lives on the main site for anyone to read and run. The enterprise service is designed, deployed and operated with you, under contract.

Open, on the main site

  • Class assumptions
  • Power and thermal methods

Enterprise, on this site

  • MEC deployment
  • Edge AI sites

06The rest of the line

Edge compute & AI, area by area.

Edge compute & AI

AI tower

A modular edge data centre, delivered as one unit.
Edge compute & AI

MEC

Applications one hop from the radio.
Edge compute & AI

Edge AI

Inference on the stream, not the archive.
Edge compute & AI

Physical AI

Drones and robots, inside published red lines.

Bring us the site.

Survey, design, deployment and operation for scientific compute. The form names who receives your details.

Talk to uscontact@observatory.global

Run the model where the data lands.

Scientific compute, inside Edge compute & AI: EO foundation models, training data and labels, land cover and change detection, hazard and exposure mapping and more. Every item is traced to its specification, from OGC Training Data Markup Language for AI (TrainingDML-AI) to CEOS Analysis Ready Data framework, and runs on the open Observatory Protocol.

01What it is

02Architecture

Where it sits, and what it speaks.

03On the grid

04In the field

Where this runs, sector by sector.

05Open and enterprise

Open by default. Enterprise by contract.

The open counterpart lives on the main site for anyone to read and run. The enterprise service is designed, deployed and operated with you, under contract.

Open, on the main site
    Enterprise, on this site
      06The rest of the line

      Bring us the site.