Bimanual Manipulation
Bimanual manipulation research focuses on enabling robots to perform tasks requiring coordinated use of two arms, mirroring human dexterity. Current efforts concentrate on developing robust control policies through imitation learning, often employing hierarchical attention transformers, graph neural networks, and model predictive control to manage the high-dimensional action space inherent in such tasks. This field is crucial for advancing robotics in areas like surgery, manufacturing, and daily assistance, with recent work emphasizing the development of efficient data collection methods and the creation of comprehensive benchmarks for evaluating progress. The ultimate goal is to create robots capable of performing complex, real-world tasks that demand coordinated two-handed manipulation.
Papers
Bi-Touch: Bimanual Tactile Manipulation with Sim-to-Real Deep Reinforcement Learning
Yijiong Lin, Alex Church, Max Yang, Haoran Li, John Lloyd, Dandan Zhang, Nathan F. Lepora
BiRP: Learning Robot Generalized Bimanual Coordination using Relative Parameterization Method on Human Demonstration
Junjia Liu, Hengyi Sim, Chenzui Li, Fei Chen