Bridging the Sim2Real Gap: Vision Encoder Pre-Training for Visuomotor Policy Transfer
Biruduganti, Samuel, Yardi, Yash, Ankile, Lars
–arXiv.org Artificial Intelligence
Simulation offers a scalable and efficient alternative to real-world data collection for learning visuomotor robotic policies. However, the simulation-to-reality, or "Sim2Real" distribution shift -- introduced by employing simulation-trained policies in real-world environments -- frequently prevents successful policy transfer. This study explores the potential of using large-scale pre-training of vision encoders to address the Sim2Real gap. We examine a diverse collection of encoders, evaluating their ability to (1) extract features necessary for robot control while (2) remaining invariant to task-irrelevant environmental variations. We quantitatively measure the encoder's feature extraction capabilities through linear probing and its domain invariance by computing distances between simulation and real-world embedding centroids. Additional qualitative insights are provided through t-SNE plots and GradCAM saliency maps. Findings suggest that encoders pre-trained on manipulation-specific datasets generally outperform those trained on generic datasets in bridging the Sim2Real gap. https://github.com/yyardi/Bridging-the-Sim2Real-Gap
arXiv.org Artificial Intelligence
Jan-25-2025
- Country:
- North America > United States
- Illinois (0.04)
- Massachusetts > Middlesex County
- Cambridge (0.04)
- North America > United States
- Genre:
- Research Report > New Finding (0.88)
- Technology:
- Information Technology
- Sensing and Signal Processing > Image Processing (1.00)
- Artificial Intelligence
- Vision (1.00)
- Robots (1.00)
- Machine Learning (1.00)
- Information Technology