Webinar Details
Training a robot to navigate in simulation is one thing. Getting that same policy to perform reliably on real hardware is where most projects hit a wall. In this session, Quanser R&D Engineer Zinan Cen pulls back the curtain on an active QBot mobile robot navigation project, walking through the methodology, mid-project discoveries, and open challenges involved in closing the sim-to-real gap. Rather than a polished case study, this is a candid look at the diagnostic process behind an in-progress reinforcement learning pipeline, built on Isaac Lab and deployed to real hardware: what’s worked, what hasn’t, and what the team is still figuring out. Ideal for robotics engineers, reinforcement learning researchers, and educators building or deploying learning-based robotics systems.
What You’ll Learn
- A framework for thinking about where sim-to-real gaps come from and how to start diagnosing them
- Real examples of mid-project discoveries, including a few that didn’t go as planned
- An inside look at the tools and workflow behind an active Isaac Lab to real-hardware RL pipeline
Presenter’s Bio
Zinan Cen
Zinan is an R&D Engineer in Academic Applications at Quanser, where he works on robotics and reinforcement learning projects spanning simulation-to-real-world deployment. He holds a master’s degree in Robotics and an undergraduate degree with a minor in Robotics from the University of Toronto. At Quanser, he focuses on developing and refining sim2real methodologies for mobile robot and manipulator platforms, bridging simulation-based training with real hardware deployment for academic applications.
Webinar Registration
Please note that your information is saved on our server as you enter it.
