The Reality Gap in Robotics
Simulation is a cornerstone of modern robotics, offering a safe, scalable, and efficient environment for training and testing. Develop and validate complex behaviors without risking real hardware. Compare algorithms in reproducible, controlled scenarios. Create vast datasets for learning at a fraction of the real-world cost. However, the effectiveness of simulation is often limited by the "reality gap"—the discrepancies between simulated and real-world environments. Simulations are only approximations of the real world. The "reality gap" is the collection of discrepancies—from physics to perception—that can cause a policy to fail when transferred from simulation to a real robot. This survey provides a comprehensive overview of the sim-to-real landscape. We dissect the problem, identify the sources of the reality gap, and provide metrics and solutions to understand and alleviate these problems in practice. Our goal is to boost the understanding of the problem by providing a guide for re
The Reality Gap in Robotics: Challenges, Solutions, and Best Practices 1University of Zurich 2NVIDIA 3The University of Sydney 4University of Washington 5University of Utah Overview The Power of Simulation in Robotics Simulation is a cornerstone of modern robotics, offering a safe, scalable, and efficient environment for training and testing. However, the effectiveness of simulation is often limited by the "reality gap"—the discrepancies between simulated and real-world environments. The Reality Gap Simulations are only approximations of the real world. The "reality gap" is the…
saved by
related reading
- Robotic World Model: A Neural Network Simulator for Robust Policy Optimization in Roboticsarxiv.org
- Robots Must Be Ephemeralizedevjang.com
- Building Worlds That Train Robotsworldlabs.ai
- Stop Simulating, Start Experiencingpaoloai.substack.com
- Eric Jang: Robots Must Be Ephemeralizedblog.evjang.com
- State of Robot Learning, December 2025vedder.io
- Overcoming the Sim-to-Real Gap: Leveraging Simulation to Learn to Explore for Real-World RLarxiv.org
- Behavioral cloning mysteryseohong.me
- A Functional Taxonomy of World Models - Dr. Fei-Fei Lidrfeifei.substack.com
- How Can We Make Robotics More like Generative Modeling? | Eric Jangevjang.com
- Domain Randomization for Sim2Real Transfer | Lil'Loglilianweng.github.io
- Sporks of AGIsergeylevine.substack.com