Visual Intelligence Summit: Oct 22 in San Francisco Get your ticket (opens in a new tab)

Roboflow + Ignition

Write detections, counts and pass or fail results from Roboflow Workflows into Ignition tags over OPC UA or MQTT, for Perspective screens, history and alarms.

Ignition by Inductive Automation is an industrial platform for SCADA, HMI and MES, built around a Gateway that holds tags and serves Perspective screens.

Roboflow writes Workflow results into those tags in two ways. The OPC UA Writer Sink block writes a boolean, integer, float or string to a tag on Ignition's own OPC UA server, using direct NodeId lookup for Ignition's string-based tag paths. Or the MQTT Writer block publishes to Ignition's MQTT Distributor broker, and the MQTT Engine module turns each topic into a tag. Either way, results land next to the PLC data Ignition already collects.

How to connect Ignition to Roboflow

5 steps
  1. Set up an edge device

    Run your Workflows on a device managed by Deployment Manager, or on a self-hosted Inference server with enterprise blocks, on a network that can reach the Ignition Gateway.

  2. Expose Ignition's tags over OPC UA

    With the OPC UA module installed, open the Gateway's OPC UA server settings and turn on Expose Tag Providers, so outside OPC UA clients can write to your tags.

  3. Create tags for the results

    In Ignition Designer, add a tag for each value the Workflow will send, such as a Boolean reject flag or an Integer defect count, and note its NodeId.

  4. Add an OPC UA Writer Sink for each tag

    In the Workflow, add one OPC UA Writer Sink per tag. Enter Ignition's OPC UA endpoint URL, set Node Lookup Mode to direct, enter the tag's NodeId and value type, and add a username and password if the server requires them.

  5. Use the tags in Ignition

    Bind the tags to Perspective components, turn on history to keep a record, and add alarms so operators hear about defects as they happen.

What teams build with Ignition

  • Show inspection results in Perspective

    Put live pass or fail status, defect counts and confidence on the screens operators already watch.

  • Log counts to the historian

    Turn on history for a count tag to chart output and defects by shift alongside other line data.

  • Raise an alarm on a defect

    Set an Ignition alarm on a reject flag so the right people are notified the moment a defect is found.

Questions

OPC UA writes straight to a tag on Ignition's own server with no extra modules. MQTT suits sites that already run the MQTT Distributor and MQTT Engine modules, or that send data through one broker to several systems.

Turn on Expose Tag Providers in the OPC UA server settings so outside clients can reach your tags. In the OPC UA Writer Sink, use direct lookup mode with the tag's NodeId, since Ignition uses string-based tag paths.

No. This integration sends results from Roboflow into Ignition, and the OPC UA Writer Sink writes values without reading them.

No. The OPC UA and MQTT blocks run on your own device or self-hosted Inference server, which needs a network route to the Ignition Gateway.

The OPC UA Writer Sink and MQTT Writer are part of the Enterprise plan. Talk to our team to add them to your workspace.

Why are companies choosing Roboflow?

The end-to-end platform for vision AI: create datasets, train models and deploy to production.

  • Flexible deployment modes

    Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.

  • Integrated workflow builder

    Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.

  • Model monitoring and alerts

    Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.

View all
Get started

Build on the Platform

For developers, engineers, and technical founders who want to get hands on. Try the free tier; the docs are open.

Bring it into your operation

For heads of AI, operations leaders, and enterprise teams. Bring a known problem, or work with us to find the first one worth taking on.