Solution visualizationLoRaWAN farm monitoring
Bring distributed sensor readings into one application.
View the solutionSOLUTIONS / EDGE APPLICATIONS & AI
Edge computing & AI integration
Combine field data with local application services. For vision or AI projects, evaluate the model, input streams and processing requirements on the target platform.
Build a project briefExplore the application
See the setting and the system.
Select a scenario to inspect the architecture, expected outputs and installation requirements.
Application examples
How this fits your site
Each example connects the field equipment, the edge system and the next operational step. Configuration and delivery are agreed around your installation.
Application example
On-site data transformation & workflows
For integrators combining industrial data with application-specific rules and message formats.
Discuss this application- Modbus / MQTT data
- Node-RED workflow
- Customer application
Use a supported MT7981 configuration or an RK3566 application computer to transform values, route events and connect agreed services through a project flow.
Project deliverables
- Approved flow and required nodes
- Runtime and configuration baseline
- Startup and persistence tests
To configure your projectRequired nodes, data rate, memory, storage and workflow ownership.
Project integration
Local monitoring applications
For sites needing an application beside the equipment, with local service access and a defined upstream interface.
Discuss this application- Equipment data
- Ubuntu application computer
- Local UI / upstream system
Integrate the agreed Ubuntu service with field data and local access. Specify storage, retention and restart behavior for the actual application rather than relying on a continuous cloud session.
Project deliverables
- Application deployment configuration
- Storage and service-lifecycle plan
- Local-access and recovery tests
To configure your projectApplication stack, resource use, retention and deployment method.
Project evaluation
Edge vision for remote sites
For an integrator evaluating camera events alongside equipment status at a pump site, utility room or machine installation.
Discuss this application- Camera stream
- RK3566 + selected vision model
- Event metadata / operator review
Evaluate a defined camera stream and model on the target hardware. Connect agreed event metadata to the local workflow and validate detection behavior using representative site footage.
Project deliverables
- Camera and model configuration
- Workload and event-integration record
- Site-data evaluation report
To configure your projectCamera, model, lighting, required events, latency and storage.
Application requirements
Functions and interfaces
An integrator needs local processing alongside equipment data, with a defined path from an application trial to a deployable configuration.
- Run local application services on Ubuntu
- Build and connect Node-RED workflows
- Evaluate project-specific NPU-assisted vision integration
Ubuntu · Node-RED · application-specific AI integration
Related products
Product families for this application
Integration
Configuration and acceptance scope
AI integration is evaluated per project. Model accuracy, camera counts, latency and unattended operation are not universal product guarantees.
Prepare the requirementsFind the right approach for your site
Share your equipment, interfaces and project goals. Discuss a standard model, an integration pilot or a small-batch custom system.