Leveling up manufacturing with virtual reality and AI
Researchers team up with Los Alamos National Laboratory and ASU’s SWAP Hub to create immersive training simulations
ASU prep academy student tests out SWAP Hub’s Future Fab Heroes desktop game. Photo courtesy of Cameron Burridge/ASU
Picture this: a virtual learning factory where machines hum and controls respond just like the real thing, but without the real risks. With digital twins, advanced manufacturing training is evolving beyond textbooks and lectures and into immersive gaming experiences.
In fields like semiconductor fabrication and nuclear materials handling, even small mistakes can have serious consequences. These hazardous, tightly regulated environments require adaptable training to ensure safety and compliance. Unfortunately, traditional training methods can fail to reflect the complexity and precision of advanced manufacturing.
Since semiconductors are such a specialized topic, they’re often left out of K-12 education, leaving many young learners unaware of their existence, relevance or real-world applications. That’s why it’s important to spark student interest from an early age.
Postdoctoral researchers and game programmers in the School of Manufacturing Systems and Networks, part of the Ira A. Fulton Schools of Engineering at Arizona State University, are exploring how virtual reality, or VR, and artificial intelligence, or AI, technologies can be used to improve industry training methods and address K-12 awareness gaps. The goal is to simulate real-world scenarios and transform how skills are introduced, taught and mastered.
An image captured from a vertical vantage point showing the digital operator instructor alongside items commonly used in a glovebox lab, positioned within a simulated storage environment. Photo courtesy of Harlan Sturzenegger/ASU
There are two sides to the ASU’s approach. One focuses on industry, where researchers are developing the next wave of workforce training tools — with resources from Los Alamos National Laboratory, or LANL — to help people build the skills needed for high-tech manufacturing roles. The other centers on K-12 education, using game-based learning to introduce students to semiconductors at an early age, supported through funding from the CHIPS Act.
Handle with care: Why train in VR?
The VR collaboration between ASU and LANL focuses on customized training for glovebox operations.
A glovebox is a sealed container designed to prevent chemical or radiological contamination. Technicians handle materials inside it using built-in gloves, allowing them to work safely without risk of direct exposure.
“Everyone at Los Alamos National Laboratory spends the first four to five months just on glovebox training,” explains Akshay Ganesh Bharadwaj, a postdoctoral scholar in the School of Manufacturing Systems and Networks assigned to the LANL glovebox project.
“We’re hoping to create VR simulations that efficiently teach muscle memory, hand-eye coordination and proper glovebox scanning techniques,” he says.
These immersive environments will go beyond basic technical training, providing a safe space to practice the specific motions and processes essential to ensure timeliness and quality control.
“VR allows us to replicate the environment and mimic procedures safely and repeatedly,” Bharadwaj says.
The VR simulation guides learners through exact sequences, such as checking pressure gauges, scanning areas with radiation detectors and contamination detection. Visual cues such as red outlines and grids help learners follow the correct scanning speed and coverage. To replicate real instructor guidance, the team integrated AI-driven voice systems.
“If contamination is detected, the AI interprets the user’s next command, advancing the scenario,” says Harlan Sturzenegger, assistant game programmer in the School of Manufacturing Systems and Networks. “All voice guidance is generated dynamically using text-to-speech and speech-to-text technology.”
This conversational AI element allows trainees to practice realistic communication and decision-making without a human instructor present.
Training Future Fab Heroes
Running parallel to the LANL project, ASU’s team in the Southwest Advanced Prototyping Hub, or SWAP Hub, is developing interactive gaming modules aimed at younger students, designed to build early interest in semiconductor careers.
Cameron Burridge, assistant game programmer in the School of Manufacturing Systems and Networks, is leading this push.
Future Fab Heroes is an immersive desktop, mobile and tablet game hosted on the platform Spatial that the team is developing for classroom integration.
“Future Fab Heroes is set on Mars in the year 3000,” Burridge says. “Students are guided by a friendly robot named CHIP-E who introduces them to fabrication processes through a series of interactive mini-games and challenges.”
The educational game teaches core semiconductor principles such as photolithography, wafer creation and proper cleanroom gowning through easy-to-understand lessons.
“Students play games, earn badges and take quizzes along the way,” Burridge says. “All narration is AI-generated to ensure consistent, accessible directions tailored for young learners.”
Designed with a futuristic theme, the game is meant to ignite curiosity and introduce core concepts in an engaging, accessible format. Early feedback has been promising.
Surveys show that about 40% of participants reported increased interest in semiconductor careers, with the strongest gains among those who were previously undecided or indifferent. This suggests the game is especially effective at engaging open-minded but less informed students. No participants reported a decline in interest, and 85% of students were open to a follow-up experience.
ASU student game developers test out the Future Fab Heroes game. Photo courtesy of Cameron Burridge/ASU
From classroom to cleanroom
A preliminary pilot at ASU Preparatory Academy gave middle and high school students an opportunity to test this game, offering early insights into its potential educational impact. Results showed a marked improvement in skill acquisition and retention compared to traditional methods, prompting plans for a broader rollout this September at the Arizona Science Center.
The team’s long-term plan is to integrate this experience into classrooms, complete with a fully-supported curriculum. Meanwhile, LANL’s VR training is moving closer to official integration.
“We’ve held informal testing with younger users to improve the interface and AI,” Bharadwaj explains. “The goal is to release a deployable version by next year so LANL can start onboarding trainees and compare new results to traditional training.”
Virtual reality and gaming are becoming a powerful learning tool for students and industry alike.
“We have check-ins every four months,” Sturzenegger adds. “Eventually, trainee performance data will feed into the learning management system, giving credit for hours and skills, similar to platforms like Canvas.”
Though these two projects serve different groups, they both use immersive technology that’s becoming more dynamic and personalized.
“We’re building training tools that will continue to progress as technology evolves,” Bharadwaj says.
