VECTORIO“Sensor feeds, contested links, no cloud, and results nobody can reproduce”
Fuse at the edge. Prove every result.
Same fusion logic at the edge and in the rear, comparable results, no rewrite per tier
- Every sensor speaks a different protocol and every integration is a one-off project.
- I cannot ship raw sensor data over a contested link, but nobody lets results leave unproven.
- The demo ran on a cluster. My platform is a rugged box with no network.
- Same fusion logic at the edge and in the rear, comparable results, no rewrite per tier
- Less raw data over contested links; heterogeneous feeds integrated once
- Reconstructable after-action picture; evidence travels with the result
From today to the first result.
a one-off integration for every sensor
Each sensor protocol is its own project, and after an incident the fusion pipeline cannot be replayed exactly.
One deterministic workflow from cluster to a Raspberry Pi. Processing at the source runs at 18.7 ms per frame at 30 fps with 0 % drops, and any incident can be replayed.
Connect one sensor stream to a workflow on a spare box beside the existing system.
one sensorLive camera, on-device detection. Step by step.
A live camera stream is analysed by a local model; only confident detections are kept and sealed as evidence.
- CameraRTSP Stream Input
- Detect · MonitorONNX Classifier · Video Viewer
- Row FilterETL.Filter
- EvidenceEvidence Collector
- PackageEvidence Pack
camera-input → Video.RTSP.Subscriber
object-detect → Classification.Onnx
confidence-filter → ETL.Filter
evidence-collect → Forensic.EvidenceCollector
evidence-pack → Forensic.EvidencePack
viewer-output → Video.Viewer- CameraReads a live camera stream.
- DetectRuns an ONNX model on each row or image.
- MonitorShows frames live in a window.
- Row FilterKeeps only rows that satisfy a condition.
- EvidenceHashes rows and batches into a tamper-evident evidence manifest.
- PackageSeals a manifest into a verifiable evidence package.
A real workflow from our end-to-end test suite; the designer view is illustrated.
What you get. From the first pilot.
Same fusion logic at the edge and in the rear, comparable results, no rewrite per tier
Less raw data over contested links; heterogeneous feeds integrated once
Reconstructable after-action picture; evidence travels with the result
Why your tools fall short.
Cloud-first platforms and managed streaming assume connectivity and a vendor cloud; they fail air-gapped or DDIL (disconnected, intermittent, limited) conditions.
Heavy stacks (Spark/Flink/Kafka clusters) do not fit a vehicle, UAV ground station or Raspberry-Pi-class node; teams rewrite logic per tier.
Point-to-point integrators and bespoke middleware are non-deterministic and carry no replayable audit trail.
- European Defence Fund (EDF) 2021-2027: nearly EUR 7.3 bn, EUR 2.7 bn collaborative research and EUR 5.3 bn capability development (EC EDF page)
- EUDIS (European Defence Innovation Scheme) and EDF work programmes
- NATO interoperability / federated mission networking
Your pilot. Three steps.
Your data shape
Tell us what data you have, where it lives and what the result must be.
A scoped pilot
On your data and your hardware, with the success measure written down first.
A clear decision
You see the result against that measure — and keep every workflow and record.
Related roles in Sovereign & edge.
Bring us your hardest data problem. Let's solve it together.
A short pilot on your data and your hardware. We agree on what success looks like before we start.