The “Teach Industrial Robotics with AI and XR” (TI-RAX) project represents a leap forward in modern industrial education. Developed to address the growing need for specialized skills in advanced manufacturing, the primary purpose of TI-RAX is to enable learners to acquire practical, hands-on competencies for safe and effective collaboration with robotic systems.
TI-RAX is a pedagogical methodology linked to a technological framework: as such, it is exploitable in multiple and diverse scenarios. To demonstrate the value of these concepts, TI-RAX implemented two highly representative industrial scenarios: sorting wasted automotive batteries and dismantling refrigerators.
Pedagogical methodology
Learning experiences are structured as guided, decision-driven simulations in shared human-robot collaboration workspaces. Through immersive extended reality, trainees understand safety procedures, environmental risks, and operational aspects of robot-assisted operations. The material emphasizes hazard identification, the proper use of protective devices, safe interaction zones, diagnostic feedback, and making critical decisions.
The curriculum supports vocational education and reskilling pathways aligned with the European qualifications framework (EQF) level 6. Also, the process is designed to be compatible with micro-credential frameworks (such as open badges on BESTR), supporting traceable learning outcomes and modular certification.
Technical methodology and implementation
TI-RAX employs a highly innovative methodology centered around an intelligent multi-agent generation pipeline. This system acts as an automated co-designer, translating structured educational specifications—in standard text—directly into functional, VIROO-ready source code. This rapid prototyping capability ensures tight alignment between the pedagogical intent and the final technical implementation.
Figure 1: TI-RAX pipeline.
Figure 1 shows the technical pipeline from the textual description of a scenario to its translation in Unity™ code for the VIROO™ platform.
Results and validation
Validation revealed high completion rates and high task correctness in the guided instructional modules. Autonomous operations where learners act without guidance proved to be more challenging, suggesting the need for adaptive scaffolding during the autonomous modules. Qualitative feedback praised the immersive realism and the structured transition from theory to practice.
Giving back to science
TI-RAX embraces a strong open-science philosophy. All the materials are available via open-source repositories on GitHub. In addition, the team has expanded the project’s reach through social media campaigns, business matchmaking events, the AI4Manufacturing network, presentation at the ITAL-IA ’26 (done) and at SIREM Engaging Education workshop (November 2026). Other scientific dissemination actions are planned well beyond the end of TI-RAX.