A vision-language conditioned physics-aware imitation learning approach for bimanual robotic dexterous assembly

Keyphrases

• Learning Approaches 100%
• Robotics 100%
• Bimanual 100%
• Imitation Learning 100%
• Physics-aware 100%
• Vision-language 100%
• Collaborative Assembly 66%
• Coarse-to-fine 33%
• Task Planning 33%
• Assembly Task 33%
• Increasing Complexity 33%
• Dual-arm 33%
• Language Data 33%
• Hyperreality 33%
• Mixed Reality 33%
• Dexterous Hand 33%
• Skill Learning 33%
• Robotic Platform 33%
• Vision Information 33%
• Tactile Information 33%
• Chain-of-Thought 33%
• Vision-Language Models 33%
• Manipulation Skill 33%
• Bilateral Teleoperation System 33%
• Contact-rich Manipulation 33%
• Diffusion Policy 33%
• Hand-based 33%

Computer Science

• Robotics 100%
• Learning Approach 100%
• Collaboration 100%
• Learning Skill 50%
• Robot 50%
• Mixed Reality 50%
• Language Modeling 50%
• Tactile Information 50%
• Chain-of-Thought 50%