Visit Today : 41
Visit Yesterday : 143
This Month : 2407
This Year : 16038
The Conference on Computing in High Energy and Nuclear Physics (CHEP) serves as a major forum for addressing the software and computing challenges of large-scale experiments in high-energy physics. The 2024 event was jointly hosted by AGH University of Kraków, the Institute of Nuclear Physics Polish Academy of Sciences, and Jagiellonian University.
Held in Kraków, CHEP 2024 continued the conference’s tradition of rotating between the Americas, Asia, and Europe. This year’s edition featured a packed program of scientific sessions, technical contributions, and community engagement.
On behalf of the ATLAS Collaboration at CERN, a team of researchers from Georgian Technical University has developed an innovative solution to make virtual reality (VR) experiences in high-energy physics education accessible on low-cost mobile devices.
The project was led by Alexander Sharmazanashvili from GTU’s Nuclear Engineering Center, and the results were presented at the Computing in High Energy Physics (CHEP) 2024 conference, earning recognition from the international scientific community.
Traditional VR tours of complex particle detectors like ATLAS typically require expensive hardware—such as Oculus headsets, Microsoft HoloLens, and high-end computers—making them impractical for widespread educational use.
“Our goal was to bring high-quality VR experiences to everyone, including those with only a basic smartphone and a Google Cardboard,” the team explains. “This opens up access to world-class physics facilities for students and educators around the globe.”
A major obstacle was the geometry complexity of the detector itself. The full “as-built” geometry of the ATLAS detector contains approximately 45 million solid geometry primitives, resulting in tens of billions of triangles—far beyond what mobile devices can process in real time.
To solve this, the team developed a seven-step geometry simplification method to drastically reduce computational load while preserving visual and educational value:
Parameter Definition – Setting screen resolution and visibility thresholds
Component Removal – Excluding elements too small to be seen
Parts Removal – Removing non-essential structures for educational use
Hole Removal – Eliminating geometry gaps that inflate triangle counts
Profile Transformation – Replacing curved shapes with straight-line alternatives
Surface Simplification – Merging repetitive features into simpler forms
Approximation Optimization – Balancing detail and performanceThe method achieved dramatic reductions in geometry complexity. In one test case of the ATLAS Muon Barrel region, the model was reduced from nearly 57 million triangles to just 94,481, a reduction of over 99%—all while retaining educational clarity.
The simplified models are now integrated into Tracer/VR (https://tracer-vr.web.cern.ch), a browser-based VR application that runs smoothly on average smartphones with Google Cardboard. It features gyroscope-based navigation and provides virtual tours of ATLAS detector components.
This breakthrough allows schools, universities, and science museums around the world to offer immersive VR tours without the cost barrier of high-end hardware. It marks a significant step forward in democratizing science education through technology.
The work also showcases Georgia’s growing role in cutting-edge educational tools for global science outreach.
The research team includes:
Alexander Sharmazanashvili (GTU, project lead)
Roger Jones (Lancaster University)
Alexander Alikhanov
Giorgi Mirziashvili
Kote Tsutskiridze
Ela Abramovi
Avtandil Khelashvili
—all affiliated with Georgian Technical University’s Nuclear Engineering Center.
