Device control / machine learning inference at edge
Traditionally, the manufacturing industry has relied on PLCs and industrial software like SCADA / DCS / MES running on-premises for device control and process orchestration or automation. The industry is increasingly adopting cloud technologies to augment these local capabilities.
AI/ML at the edge is one such augmentation. AWS provides a set of tools that make AI/ML readily accessible to any organization. Manufacturers can utilize these advanced tools to solve process control challenges. They can train the model in the cloud and deploy it on the edge to leverage ML for advanced process control. For example, customers can add visual inspection monitored by AI/ML to improve the detection of defects and exceptions.
Process orchestration and control using AWS IoT Greengrass
Example of process orchestration with AWS IoT Greengrass
FreeRTOS
In this scenario, the cloud-enabled component could exist in Levels 0-3 of the plant networks. With the ability to write back to the controllers and control industrial equipment, this scenario warrants careful security planning and implementation.