# Learning Dual-Arm Coordination for Grasping Large Flat Objects

arXiv: https://arxiv.org/abs/2504.03500 · HTML: https://mlsanity.com/paper/2504.03500/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 17 of its 30 references that are in the map.
- First posted on arXiv 4 Apr 2025.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 17 of 17)

- [Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actor](https://mlsanity.com/paper/1801.01290/) (2018)
- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [AnyGrasp: Robust and Efficient Grasp Perception in Spatial and Temporal Domains](https://mlsanity.com/paper/2212.08333/) (2022)
- [Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learning](https://mlsanity.com/paper/1803.09956/) (2018)
- [Deep Learning Approaches to Grasp Synthesis: A Review](https://mlsanity.com/paper/2207.02556/) (2022)
- [Deep Visual Reasoning: Learning to Predict Action Sequences for Task and Motion Planning from an Initial Scene Image](https://mlsanity.com/paper/2006.05398/) (2020)
- [Goal-conditioned dual-action imitation learning for dexterous dual-arm robot manipulation](https://mlsanity.com/paper/2203.09749/) (2022)
- [Towards Scale Balanced 6-DoF Grasp Detection in Cluttered Scenes](https://mlsanity.com/paper/2212.05275/) (2022)
- [Efficient learning of goal-oriented push-grasping synergy in clutter](https://mlsanity.com/paper/2103.05405/) (2021)
- [Learning Pregrasp Manipulation of Objects from Ungraspable Poses](https://mlsanity.com/paper/2002.06344/) (2020)
- [Attribute-Based Robotic Grasping with Data-Efficient Adaptation](https://mlsanity.com/paper/2501.02149/) (2025)
- [Self-Supervised Learning for Joint Pushing and Grasping Policies in Highly Cluttered Environments](https://mlsanity.com/paper/2203.02511/) (2022)
- [A Coarse-to-Fine Framework for Dual-Arm Manipulation of Deformable Linear Objects with Whole-Body Obstacle Avoidance](https://mlsanity.com/paper/2209.11145/) (2022)
- [Visual-tactile Fusion for Transparent Object Grasping in Complex Backgrounds](https://mlsanity.com/paper/2211.16693/) (2022)
- [DenseTact-Mini: An Optical Tactile Sensor for Grasping Multi-Scale Objects From Flat Surfaces](https://mlsanity.com/paper/2309.08860/) (2023)
- [Throwing Objects into A Moving Basket While Avoiding Obstacles](https://mlsanity.com/paper/2210.00609/) (2022)
- [Learning Bifunctional Push-grasping Synergistic Strategy for Goal-agnostic and Goal-oriented Tasks](https://mlsanity.com/paper/2212.01763/) (2022)

## Abstract (arXiv)

Grasping large flat objects, such as books or keyboards lying horizontally, presents significant challenges for single-arm robotic systems, often requiring extra actions like pushing objects against walls or moving them to the edge of a surface to facilitate grasping. In contrast, dual-arm manipulation, inspired by human dexterity, offers a more refined solution by directly coordinating both arms to lift and grasp the object without the need for complex repositioning. In this paper, we propose a model-free deep reinforcement learning (DRL) framework to enable dual-arm coordination for grasping large flat objects. We utilize a large-scale grasp pose detection model as a backbone to extract high-dimensional features from input images, which are then used as the state representation in a reinforcement learning (RL) model. A CNN-based Proximal Policy Optimization (PPO) algorithm with shared Actor-Critic layers is employed to learn coordinated dual-arm grasp actions. The system is trained and tested in Isaac Gym and deployed to real robots. Experimental results demonstrate that our policy can effectively grasp large flat objects without requiring additional maneuvers. Furthermore, the policy exhibits strong generalization capabilities, successfully handling unseen objects. Importantly, it can be directly transferred to real robots without fine-tuning, consistently outperforming baseline methods.