analiza aiqbench
EquiBim addresses a critical gap in robotic imitation learning by explicitly enforcing bilateral symmetry constraints to ensure consistent dual-arm behaviors under symmetric observations. By formulating physical symmetries as group actions, this framework provides an effective inductive bias that improves performance and robustness across diverse simulation and real-world configurations without requiring model-specific changes. This advancement represents significant progress in Machine Learning for robotics, specifically targeting the structural integrity of policies in bimanual manipulation tasks.
EquiBim: Learning Symmetry-Equivariant Policy for Bimanual Manipulation
arXiv:2603.08541v3 Announce Type: replace Abstract: Robotic imitation learning has achieved impressive success in learning complex manipulation behaviors from demonstrations. However, many existing robot learning methods do not explicitly account for the physical symmetries of robotic systems, often resulting in asymmetric or inconsistent behaviors under symmetric observations. This limitation is particularly pronounced in dual-arm manipulation, where bilateral symmetry is inherent to both the robot morphology and the structure of many tasks. In this paper, we introduce EquiBim, a symmetry-equivariant policy learning framework for bimanual manipulation that enforces bilateral equivariance between observations and actions during training. Our approach formulates physical symmetry as a group action on both observation and action spaces, and imposes an equivariance constraint on policy predictions under symmetric transformations. The framework is model-agnostic and can be seamlessly integrated into a wide range of imitation learning pipelines with diverse observation modalities and action representations, including point cloud-based and image-based policies, as well as both end-effector-space and joint-space parameterizations. We evaluate EquiBim on RoboTwin, a dual-arm robotic platform with symmetric kinematics, and evaluate it across diverse observation and action configurations in simulation. We further validate the approach on a real-world dual-arm system. Across both simulation and physical experiments, our method consistently improves performance and robustness under distribution shifts. These results suggest that explicitly enforcing physical symmetry provides a simple yet effective inductive bias for bimanual robot learning.
Abstract:Robotic imitation learning has achieved impressive success in learning complex manipulation behaviors from demonstrations. However, many existing robot learning methods do not explicitly account for the physical symmetries of robotic systems, often resulting in asymmetric or inconsistent behaviors under symmetric observations. This limitation is particularly pronounced in dual-arm manipulation, where bilateral symmetry is inherent to both the robot morphology and the structure of many tasks. In this paper, we introduce EquiBim, a symmetry-equivariant policy learning framework for bimanual manipulation that enforces bilateral equivariance between observations and actions during training. Our approach formulates physical symmetry as a group action on both observation and action spaces, and imposes an equivariance constraint on policy predictions under symmetric transformations. The framework is model-agnostic and can be seamlessly integrated into a wide range of imitation learning pipelines with diverse observation modalities and action representations, including point cloud-based and image-based policies, as well as both end-effector-space and joint-space parameterizations. We evaluate EquiBim on RoboTwin, a dual-arm robotic platform with symmetric kinematics, and evaluate it across diverse observation and action configurations in simulation. We further validate the approach on a real-world dual-arm system. Across both simulation and physical experiments, our method consistently improves performance and robustness under distribution shifts. These results suggest that explicitly enforcing physical symmetry provides a simple yet effective inductive bias for bimanual robot learning.
Accepted to IROS 2026. Project page: this https URL
Robotics (cs.RO)
arXiv:2603.08541 [cs.RO]
(or arXiv:2603.08541v3 [cs.RO] for this version)
Text integral sub licența deschisă a publicației · 240 cuvinte
Subiecte conexe
EquivDP3: A SIM(3)-Invariant Point-Cloud Encoder for Data-Efficient Humanoid Loco-Manipulation
arXiv · 30 sept. 2026
Bench2Dex: Benchmarking Visuo-Tactile Bimanual Dexterous Manipulation Across Dexterous Hands
arXiv · 16 sept. 2026
RESTA: Time-Scaling Augmentation for Data-Efficient Robot Imitation Learning
arXiv · 14 oct. 2026
SAVLA: Symmetry-Aware Vision-Language-Action Models for Robotic Manipulation
arXiv · 16 sept. 2026