cs.ROMar 9, 2026

EquiBim: Learning Symmetry-Equivariant Policy for Bimanual Manipulation

Authors: Zhiyuan Zhang, Aditya Mohan, Seungho Han, Wan Shou, Dongyi Wang, Yu She

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

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.

Figures & tables

Fig. 1: Symmetry-equivariant policy learning for bimanual manipulation. A symmetry transformation S\mathcal{S} defines a left–right exchange of the scene in the image coordinate frame, generating a symmetrically equivalent task instance. Given an observation O\mathcal{O} and its transformed counterpart S(O)\mathcal{S}(\mathcal{O}) , the shared policy π\pi is trained to produce equivariant predictions under the same transformation, i.e., π(S(O))≈S(π(O))\pi(\mathcal{S}(\mathcal{O}))\approx\mathcal{S}(\pi(\mathcal{O})) . The symmetry constraint is imposed at the action level and does not require architectural modifications.
Fig. 2: Overview of the proposed EquiBim framework. Given an observation O\mathcal{O} and its symmetrically transformed counterpart S(O)\mathcal{S}(\mathcal{O}) , where S\mathcal{S} denotes a left-right reflection defined in the image frame, both inputs are processed by a shared policy π\pi . The policy takes visual observations together with proprioceptive states as input and produces action predictions π(O)\pi(\mathcal{O}) and π(S(O))\pi(\mathcal{S}(\mathcal{O})) . A symmetry-equivariant loss Lsym\mathcal{L}_{\mathrm{sym}} enforces the predicted actions to transform consistently under S\mathcal{S} during training.
Fig. 3: Visualization of eight simulated bimanual manipulation tasks used for evaluation. Subfigures (a)-(g) are adapted from RoboTwin [ 19 ] . For each task, the left image shows the initial state and the right image shows the goal state.
 Observation + ActionBaseline+ EquiBimΔ\boldsymbol{\Delta}
DPImage + Joint34.143.6+9.5
Image + EE37.340.0+2.7
DP3Point Cloud + Joint73.577.9+4.4
Point Cloud + EE74.577.8+3.3
TABLE I: Average success rate (%) across observation–action settings on RoboTwin.
 Observation + ActionBeatClickHandoverMovePickPlaceStampPress
DPImage + Joint+22++ 0+2+23−- 11+28+3+9
Image + EE+44−- 6−- 15+9−- 13−- 2+1+4
DP3Point Cloud + Joint+12+7−- 1−- 6+4+6−- 2+15
Point Cloud + EE+4++ 5++ 1+3−- 5+9−- 1+12
TABLE II: Per-task performance gains from symmetry regularization ( Δ\Delta success rate, %).
Fig. 4: Overview of the real-world bimanual manipulation setup. Two LeRobot SO101 robotic arms operate in a shared workspace and are observed by a Logitech C920x camera. The system is evaluated on object handover and hook-hanging tasks under a top-down camera configuration.
Fig. 5: Real-world task executions on the bimanual LeRobot platform. The figure presents three tasks, namely Object Handover (Banana), Hook Hanging (Drumstick), and Hook Hanging (Toy Chicken), each depicted across three stages: initial, intermediate, and final. Each row shows the temporal progression of the manipulation process under a top-down RGB camera view.
Fig. 6: Generalization under distribution shift on the real-world bimanual platform. Rollouts are shown under the training distribution (left) and shifted conditions (right) for object handover, drumstick hook hanging, and toy chicken hook hanging. Each visualization overlays multiple rollouts (out of 10 total trials per setting) using transparency to illustrate trajectory variability. For clarity, only a subset of representative rollouts is displayed. The shifts include symmetric orientation changes, demonstration quality variation, and randomized initialization.
MethodTraining DistributionShifted Distribution
 BananaDrumstickToy ChickenBananaDrumstickToy Chicken
ACT3/103/108/100/101/104/10
ACT + EquiBim6/103/109/105/104/106/10
TABLE III: Success rate in real-world experiments under training and shifted object distributions.

Explore similar work

CardsList
May 12, 2026

Morphologically Equivariant Flow Matching for Bimanual Mobile Manipulation

Mobile manipulation requires coordinated control of high-dimensional, bimanual robots. Imitation learning methods have been broadly used to solve these robotic tasks, yet typically ignore the bilateral morphological symmetry inherent in such systems. We argue that morphological symmetry is an underexplored but crucial inductive bias for learning in bimanual mobile manipulation: knowing how to solve a task in one configuration directly determines how to solve its mirrored counterpart. In this paper, we formalize this symmetry prior and show that it constrains optimal bimanual policies to be ambidextrous and equivariant under reflections across the robot's sagittal plane. We introduce a C2\mathbb{C}_2-equivariant flow matching policy that enforces reflective symmetry either via a regularized training loss or an equivariant velocity network. Across planar and 6-DoF mobile manipulation tasks, symmetry-informed policies consistently improve sample efficiency and achieve zero-shot generalization to mirrored configurations absent from the training distribution. We further validate this zero-shot generalization capability on a real-world manipulation task with a TIAGo++ robot. Together, our findings establish morphological symmetry as an effective, generalizable, and scalable inductive bias for ambidextrous generative policy learning.

Max Siebenborn, Daniel Ordoñez Apraez, Sophie Lueth +4

PEARL Lab, Dept. of Computer Science, TU Darmstadt, Germany · 2Istituto Italiano di Tecnologia, 3Hessian.AI, 4Robotics Institute Germany

Flow MatchingMobile ManipulationBimanual Robotic ManipulationGeneralization in Robotic ManipulationRoboticsRobot Imitation Learning
AskPDF
May 28, 2026

MonoDuo: Using One Robot Arm to Learn Bimanual Policies

Bimanual coordination is essential for many real-world manipulation tasks, yet learning bimanual robot policies is limited by the scarcity of bimanual robots and datasets. Single-arm robots, however, are widely available in research labs. Can we leverage them to train bimanual robot policies? We present MonoDuo, a framework for learning bimanual manipulation policies using single-arm robot demonstrations paired with human collaboration. MonoDuo collects data by teleoperating a single-arm robot to perform one side of a bimanual task while a human performs the other, then swapping roles to cover both sides. RGB-D observations from a wrist-mounted and fixed camera are augmented into synthetic demonstrations for target bimanual robots using state-of-the-art hand pose estimation, image and point cloud segmentation, and inpainting. These synthetic demonstrations, grounded in real robot kinematics, are used to train bimanual policies. We evaluate MonoDuo on five tasks: box lifting, backpack packing, cloth folding, jacket zipping, and plate handover. Compared to approaches relying solely on human bimanual videos, MonoDuo enables zero-shot deployment on unseen bimanual robot configurations, achieving success rates up to 70%. With only 25 target robot demonstrations, few-shot finetuning further boosts success rates by 65-70% over training from scratch, demonstrating MonoDuo's effectiveness in efficiently transferring knowledge from single-arm robot data to bimanual robot policies.

Sandeep Bajamahal, Lawrence Yunliang Chen, Toru Lin +3

University of California, Berkeley

Bimanual Robotic ManipulationCross-Embodiment Robot LearningRobot Imitation Learning
AskPDF
Oct 12, 2026

MIRROR: Reflecting Hidden Symmetries from Offline Robot Interaction Data

Symmetry-aware robot policy learning requires coupled observation-action transformations. For pixel observations, even simple physical symmetries can induce complex, viewpoint-dependent image transformations that are difficult to specify by hand. Existing symmetry-aware methods often rely on representations with predefined transformations, requiring preprocessing tailored to the sensing setup and task. We introduce MIRROR, a framework that learns how a known symmetry group operates jointly on observations and actions in partially observable Markov decision processes (POMDPs) from offline interaction data. MIRROR couples these transformations through adversarial tuple matching over one-step transition tuples, while explicit group-consistency regularization penalizes violations of the group-operation axioms. On four Robosuite tabletop manipulation tasks, MIRROR achieves the lowest aggregate error on three of four symmetry-quality metrics, including a 61% reduction in compatibility error (which measures consistency under transformation composition) relative to an adapted LaLiGAN baseline. Using the learned transformations in frame-averaged behavior cloning (BC) yields higher mean success rates than vanilla BC on all four tasks, increasing average success from 80.48% to 85.12%, with per-task gains up to 9.66 percentage points. We also repeat symmetry discovery and policy training in simulation with heightmap and segmentation-mask observations, then deploy the resulting policy zero-shot on a real Kinova GEN3 arm. Project page: https://pomdp-mirror.github.io.

Shiqi Liu, Yihang Yao, Peide Huang +2

Carnegie Mellon University · Stanford University

Robot Policy LearningRobot Imitation LearningRobotic Manipulation
AskPDF