The Southwest Research Institute (SwRI) has introduced automation technology that enables industrial robots to visually classify work and autonomously perform tasks.

Using the technology, the SwRI team demonstrate that robots could autonomously sand and prepare surfaces on aircraft and other machinery. Additionally, the technology can be applied to grinding, painting, polishing, cleaning, welding, sealing and other industrial processes.

"Our solutions increase process repeatability while improving part quality and decreasing rework," said Matt Robinson, a robotics R&D manager at SwRI. "They also reduce human exposure to dangerous environments."

The system uses SwRI-developed machine learning algorithms and classification software that work in conjunction with open-source tools such as Scan-N-Plan and ROS 2, the latest version of the open-source robot operation system. Traditional robot programming can be slow and tedious, requiring an expert in the loop with knowledge of computer-aided design (CAD).

Scan-N-Plan, a ROS-Industrial technology, uses machine vision to scan parts, creating 3D mesh data that robots use to plan tool paths and process trajectories while performing real-time process monitoring. SwRI worked with the ROS-I project to maintain its software repository and expand open-source automation solutions.

"By leveraging these open-source tools with our custom software, we have developed a solution that intelligently classifies regions and textures of part surfaces in various stages of work," Robinson said.

The solution includes custom machine vision algorithms that enable robots to apply assorted media with varying pressures according to the amount of surface work needed. Feature-based processing is also enabled through additions that leverage semantic segmentation approaches to apply the right tool to the right feature — cutting versus sanding for instance.

"These are breakthroughs that will help prevent robots from over-sanding or over-grinding metal surfaces," said Paul Evans, director of SwRI's Manufacturing Technologies Department.