By Mark Shortt
Manufacturers today are increasingly adopting robotics to perform various operations with greater precision than humans. In manufacturing facilities across the U.S., robots are boosting the efficiency of material handling, machine tending, and assembly tasks. They’re also improving the part quality that can be achieved by production methods like machining, finishing, and laser cutting and welding.
But like talented fastball hitters getting stumped by curveballs and sliders, robots often reach their limits when they encounter situations that are anything but linear.
To eliminate production roadblocks that have long stymied robotic welding systems, some manufacturers are employing a combination of artificial intelligence, machine learning, and computer vision. By making robots smarter and more capable of adjusting to real-world part complexity and variations, this potent mix of technologies offers U.S. manufacturers the potential to enhance their production speed, adaptability, and scalability.
“Traditional automation excels at consistency but cannot adapt to the messy reality of most manufacturing environments,” according to a release from Path Robotics. The Columbus, Ohio-based company has developed an artificial intelligence (AI) model for autonomous welding that aims to solve this conundrum.
The challenges and “messy realities” that are common in manufacturing are typified by the complexity and variability encountered in high-mix, low volume welding. When a part becomes distorted from the heat of the weld, it’s like a curveball being thrown to the robotic welder, which can’t adapt to the change in weld conditions unless it’s reprogrammed or the weldment is refixtured.
The solution, according to Path Robotics, lies in making robotic welders more flexible and intelligent. Path accomplishes this through its physical AI model for welding, Obsidian™, which brings AI, machine learning, and computer vision to the welding cell, empowering it to “see, think, and adapt in real time.”
“Powered by Obsidian™, our foundational AI, Path’s Intelligent Welding Cells adapt in real-time for an agile flow of high-quality welds, even across part-to-part variations,” the company’s website states.
According to Path, Obsidian’s proprietary vision system uses lasers and optics to scan every welding seam. Data from the scan is used to develop the weld plan and path in real time.
“When welding begins, the system monitors the weld pool and seam to compensate for heat distortion as the weld progresses,” the company said in the release. “The result is consistent weld quality without the traditional burden of fixturing parts to perfection, or reprogramming every time the part changes.”
Another developer of AI software for autonomous platforms is Palladyne AI Corp., of Salt Lake City, Utah. The company’s AI and machine learning (ML) software platform—like that of Path Robotics’s—is designed to transform the capabilities of robots by “enabling them to observe, learn, reason, and act in a manner akin to human intelligence,” company executives said in a release.
Operating on the edge, the AI and ML software platform gives robots the ability to “perceive variations or changes in the real-world environment.” That information is crucial to enabling the robots to “autonomously maneuver and manipulate objects accurately in response,” the release said.
“[The platform] is designed to enable robotic systems to perceive their environment and quickly adapt to changing circumstances by generalizing (learning) from their past experience, using dynamic real-time operations on the edge without extensive programming and with minimal robot training,” the release stated.
But intelligent automation, in the form of physical AI, is not the only tool available to Palladyne AI as it strives to accelerate the production and supply of critical parts and products for the U.S. defense sector. After acquiring three U.S.-based companies in November, Palladyne AI is aiming to make its mark as a vertically integrated U.S. defense technology company that is “designed for speed, adaptability, and scalability,” according to Ben Wolff, Palladyne AI’s president and CEO.
With its recent acquisitions of GuideTech LLC, Warnke Precision Machining, and MKR Fabricators, Palladyne created a new operating division, Palladyne Defense, that unites the strengths of all three companies. The new division is envisioned as a “closed-loop innovation and production system” that supports a critical priority of the U.S. Department of Defense: the reshoring of essential defense technology and production capacity.
“Our vertically integrated model allows us to meet emerging defense and public safety needs while reshoring capability and accelerating innovation,” Wolff said in a November release. “This is how America maintains its technological advantage—by building it here, with intelligence at the edge and purpose at the core.”
Artificial intelligence at the edge of a network—that is, on an IoT device such as a robotic system—doesn’t require an internet connection and is considered more secure than AI in the cloud, which can only be accessed via the internet. A core element of its offerings, Palladyne’s embodied AI enables “edge decision-making, swarm coordination (of UAVs), and adaptive mission execution,” the release said.
Another core element of Palladyne Defense comprises GuideTech’s flight systems, avionics, weapons platforms, and rapid design-for-manufacturing services. These capabilities were developed through “accelerated prototyping and iterative engineering that compress design cycles from years to months,” according to Palladyne.
The third core element is the U.S. manufacturing and production provided by Warnke Precision Machining and MKR Fabricators. These services are reported to include “precision machining, heavy fabrication, and assembly of electronic and mechanical components that support major defense programs and Palladyne’s future proprietary systems.”
“Together, these capabilities will enable Palladyne Defense to design, develop, and produce components and complete systems entirely within the United States in the near future, strengthening supply chain security and ensuring sustained access to mission-critical components,” the company said.