features

breathing AI

Runs deep learning models on GPU for fast model inferencing.

With 21 TOPS for GPU, a 6-core ARMv8 CPU and 8GB RAM you have an AI power-engine at your disposal.

Object Detection and more

Connect multiple PoE IP cams
Stream multiple cameras in parallel
Apply object detection on all streams

Connects seamlessly to any IP cam and provides PoE for up to 8 cams.

Run object detection, anomaly detection and preventive maintenance models in parallel.

Plug-And-Play

The IIoT AI Controller seamlessly integrates with the smart IIoT Controller.

Provides full OTA functionality in combination with the smart IIoT controller.

Tailored for AI

The IIoT AI Controller is designed to run a.o. object detection solutions like the IIoT AI CAM.

At the same time it can also be used for IIoT Analytics on the edge and it can run multiple ML models in parallel.

Big performance in a small package

It includes all the functionality of a big electrical cabinet in 1,5 the size of a shoebox, eliminating the need to build costly electrical cabinets (including power supplies, PoE switch, electronic circuit breakers, switches and communication equipment), hence reducing the space of your technical area

Industrial Design

Its IP65 design and the use of M12 connectors allow you to have it installed directly in your production environment

Deep learning in real-time

The sAInce.io IIoT AI Controller is a high-end AI-ready edge device that can run the most popular DL frameworks like Tensorflow and Pytorch.

Example applications include:

Object detection

Anomaly detection

Preventive Maintenance

Industry Native

Training: is done in the cloud through AWS DL AMI and results are tracked in the cloud.
Use labelled images for object detection. Use time series data features for anomaly detection/ predictive maintenance.

Models:

Convolutional Neural Networks
XGBoost
Autoencoders
Generative Adversarial Networks
Long Short Term Memory Networks
...

Model metrics: track metrics in the cloud: mAP, precision, recall, loss, F-score etc.

generated model weights are automatically stored.

Model weights can be applied to the model service configuration

Deploy the model of choice to the AI Controller using the smart IIoT Controller's OTA service

The model runs the inferencing in real-time