EquiBim: Learning Symmetry-Equivariant Policy for Bimanual Manipulation

2026-03-09

AI Summary

This work addresses the behavioral inconsistency often observed in dual-arm robotic manipulation due to the neglect of physical symmetries. The authors propose a symmetry-equivariant policy learning framework that formalizes physical symmetries as group actions on both observation and action spaces, and incorporates equivariance constraints as inductive biases to enforce bilateral symmetry in the learned policy. The approach is compatible with multimodal observations—such as point clouds and images—and supports multiple action representations, including end-effector and joint-space commands, making it suitable for imitation learning settings. Experiments in simulation and on the real-world dual-arm platform RoboTwin demonstrate that the proposed method significantly enhances policy robustness and task consistency under distributional shifts.

Technology Category

• Intelligent Robots: Manipulation
• Machine Learning: Imitation Learning & Inverse Reinforcement Learning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

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.

Problem

• bimanual manipulation
• symmetry-equivariant
• imitation learning
• physical symmetry
• robotic policy

Innovation

• symmetry-equivariant learning
• bimanual manipulation
• imitation learning
• equivariance constraint
• robotic policy