Foceon / Platform / AI architecture
The physical-AI stack, step by step
Foceon closes a real-time loop on the shop floor. This is how perception, prediction, simulation, and action are engineered on accelerated compute so the loop runs where the glass is.
Four stages, one budget
Each stage has a latency budget so the whole loop closes fast enough to steer a live process.
- Perceivesensor fusion
- Predictlearned models
- Simulatetwin validation
- Actbounded control
How each stage is built
The architecture is deliberate about where each computation runs.
Perceive at the edge
Metrology and vision are processed at the edge on Jetson and Holoscan, so raw data becomes surface state without a round trip.
- Jetson and Holoscan at the sensor
- DeepStream vision pipelines
- Low-latency surface state
Predict and simulate on OVX
Learned models and the optics twin run on OVX and Omniverse, close enough to advise the next move in real time.
- TensorRT inference
- Omniverse twin on OVX
- Triton and NIM serving
Why it is engineered this way
Loop on the floor
The loop closes at the machine, not in a distant cloud, so it can steer a live process.
Data stays put
Perception and inference run on-prem, so proprietary optics data never leaves the fab.
Edge to cloud
The same stack scales from one Jetson to an OVX cluster without a rewrite.
Bounded action
Every action passes through interlocks and is logged against the twin.
What runs where
Metrology becomes surface state
Interferograms and profilometry are fused at the edge into a single surface state the rest of the stack reasons about.
From model to machine
Train with NeMo
Models are trained and adapted with NeMo on your process data.
Serve with Triton and NIM
Inference is served at the edge with Triton and NIM microservices.
Simulate with Omniverse
The twin validates the move on OVX before it acts.
Schedule with cuOpt
Line-level decisions are optimized with cuOpt.
Stack terms
- Holoscan
- NVIDIA's sensor-processing platform for low-latency edge pipelines.
- TensorRT
- An inference optimizer that accelerates trained models.
- Omniverse
- A simulation platform used here to run the optics digital twin.
- cuOpt
- An accelerated solver for routing and scheduling optimization.
Where the loop closes
| Property | Cloud-only AI | Foceon |
|---|---|---|
| Loop location | Remote | On the floor |
| Data locality | Leaves site | Stays on-prem |
| Real-time control | No | Yes |
| Edge-to-cloud scale | Partial | Yes |
The full NVIDIA stack
Foceon is engineered end to end on accelerated compute.
Architecture questions
Do we need NVIDIA hardware?
Foceon is engineered for it and deploys on edge and cluster configurations we scope with you.
Can it run air-gapped?
Yes. The full loop runs on-prem without external connectivity.
How is it updated?
Models and services update through controlled, auditable releases.
Put the loop on your floor
Book a pilot and we will scope the stack to your cells and metrology.