General Electric subsidiary Avitas Systems is using artificial intelligence technology from AI company Nvidia to automate industrial inspections, reducing costs, turnaround time and risk.

Using a team of drones, crawlers or wheeled robots and autonomous underwater vehicles, Avitas performs inspections at sites in the energy, transportation and oil and gas sectors, such as power plants, underwater pipelines and refineries.

Using Nvidia’s AI computing program, DGX-1, Avitas trains its network of robotics to gather data, build models and inspection paths and ultimately recognize patterns in the data. In this way, the system can detect defects or corrosion on sites and measure their severity, determining when equipment needs to be fixed or replaced.

To allow for quicker data processing, Avitas set up data centers, called AI Workbench, near inspection sites using Nvidia’s DGX Station. These centers can also retrain models to adapt to new cases.

“Working with Nvidia allows us to fully commercialize our cutting-edge, self-service AI Workbench, and we look forward to expanding its capabilities using the new Nvidia DGX Stations with Volta,” Alex Tepper, Avitas founder and head of corporate and business development, told Automation World. “With our workbench, our engineers can easily create and access new deep learning models that train the software deployed to recognize defects automatically at inspection sites.”

Automating the inspection process reduces costs and time by preventing companies from sending people out to the sites, which are often difficult to access or far away. Inspections are also completed more quickly and regularly because operations don’t need to be halted to make conditions safe for human inspectors. Avitas estimates that the new use of AI could cut costs by 25%, reduce maintenance downtown by up to 15% and increase inspection turnaround time by 25%.

And automated inspections protect the people who previously had to complete them. Instead of sending humans to inspect a petroleum refinery flare stack that operates at hundreds of degrees or a pipeline that exists deep underwater, robots can perform the task without putting employees in harm’s way or disrupting operations.

“Avitas Systems is breaking new ground by bringing Nvidia DGX Station beyond the deskside and into the field for the first time,” Nvidia general manager of DGX systems Jim McHugh said. “Using our latest DGX systems to help train robots and better predict industrial defects increases worker safety, protects the environment, and leads to substantial cost savings for companies.”