Vision and machine signals, one platform
Cameras on the line and sensors on the machine feed the same pipeline: the same dataset tools, the same training jobs, the same deployment targets.
- Image and time-series models
- Spectral, MFCC and spectrogram blocks
- Over-the-air updates
What this solves
A plant rarely has one kind of data. Vision problems and vibration or sound problems end up in separate tools, with separate deployment stories. You bring the data; the pipeline is the same for both.
Who this is for
- Automation and controls engineers
- Quality engineers
- Plant IT and OT teams
Why edge inference fits
Decisions happen locally
Inference runs on the device itself. There is no round trip to a server between a frame and a result.
Keeps working offline
A deployed model runs without a connection. Devices send heartbeats and inference logs when they reconnect.
Data stays where it is captured
Frames are processed on the device. Nothing has to leave the site for a prediction to be made.
The whole pipeline, one stage at a time
Every stage below is a screen in the platform, not a diagram of something you still have to build.
0110
01Stage 1 of 10
Collect data
Upload samples or import an existing folder tree.
Data
From the first uploaded image to a monitored device, every stage is a screen you can open.
What you get in the platform
Impulse designer
Chain a processing block and a learning block into a reproducible pipeline.
Image classification
Train MobileNetV2 from scratch or fine-tune with ImageNet transfer learning.
Sensor and audio models
Dense, 1D/2D convolutional and LSTM networks over spectral, MFCC and spectrogram features.
Deployment builds
Turn any trained model into a ready-to-run package for the target you pick.
Device fleet
Register devices with an API key, then track heartbeats and inference logs per project.
Over-the-air model updates
Resolve the compatible deployment for a device and push the update without touching the hardware.
What teams build with it
Line-side pass/fail
Classify captured frames as the product moves through the station.
Part counting
Count items with centroid detection instead of boxes.
Vibration-signature classification
Spectral analysis features into a 1D convolutional network.
Machine-sound classification
MFCC features into a classifier trained on your own recordings.
Machine state from a sensor sequence
Classify a window of readings rather than a single sample.
Catching a defect the model has not seen
Add fresh samples and retrain from the existing impulse.
Where this runs
The targets teams reach for on this kind of project.
Raspberry Pi
Python inference package
Installs with the device client and runs as a service
UNO Q
Linux runtime package
The on-device runner takes it from there, no glue code to write
TensorFlow Lite
TensorFlow Lite model with an inference library
Embed it in any runtime that can load TensorFlow Lite
Petal Edge builds for six targets in total, each with its own package format and constraints.
See all deployment optionsOne platform, start to finish
Data, labels, training, evaluation, post-processing, deployment and the fleet live in one place, with nothing to stitch together.
The whole lifecycle
Collect, label, train, evaluate, post-process, deploy and monitor without leaving the platform.
Models sized for edge hardware
NanoVision, Vision Pro and MobileNetV2 are built to run on devices, not on datacentre GPUs.
Six deployment targets
One trained model, packaged for the hardware you already run.
A fleet you can reach
Register devices, read their inference logs, and push new models over the air.
What’s included
- Dataset import from COCO, YOLO, Pascal VOC, Open Images and Edge Impulse JSON
- AI-assisted labeling with human review before anything is applied
- Confusion matrix, per-class metrics and model testing across versions
- Post-processing with NMS and IoU tracking
- Over-the-air model updates to registered devices
- Project export, so your data and models stay portable
Bring your line data
Create a project, bring your data, and take it through to a deployed model.