# Learning to Push, Group, and Grasp: A Diffusion Policy Approach for Multi-Object Delivery

arXiv: https://arxiv.org/abs/2502.08452 · HTML: https://mlsanity.com/paper/2502.08452/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Diffusion and flow policies for robotics, one of 14 directions in Diffusion models.
- Builds on 17 of its 28 references that are in the map.
- First posted on arXiv 12 Feb 2025.

## Placed in

- Diffusion models / [Diffusion and flow policies for robotics](https://mlsanity.com/diffusion/diffusion-and-flow-policies-for-robotics/)
- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

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

- [Diffusion Policy: Visuomotor Policy Learning via Action Diffusion](https://mlsanity.com/paper/2303.04137/) (2023)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [AnyTeleop: A General Vision-Based Dexterous Robot Arm-Hand Teleoperation System](https://mlsanity.com/paper/2307.04577/) (2023)
- [Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learning](https://mlsanity.com/paper/1803.09956/) (2018)
- [WiLoR: End-to-end 3D Hand Localization and Reconstruction in-the-wild](https://mlsanity.com/paper/2409.12259/) (2024)
- [Grasp Multiple Objects with One Hand](https://mlsanity.com/paper/2310.15599/) (2023)
- [Uncertainty-Aware Self-Supervised Target-Mass Grasping of Granular Foods](https://mlsanity.com/paper/2105.12946/) (2021)
- [Exploiting Kinematic Redundancy for Robotic Grasping of Multiple Objects](https://mlsanity.com/paper/2303.01868/) (2023)
- [Multi-Object Grasping -- Estimating the Number of Objects in a Robotic Grasp](https://mlsanity.com/paper/2112.01270/) (2021)
- [Self-Supervised Learning for Joint Pushing and Grasping Policies in Highly Cluttered Environments](https://mlsanity.com/paper/2203.02511/) (2022)
- [Multi-Object Grasping in the Plane](https://mlsanity.com/paper/2206.00229/) (2022)
- [Learning to Efficiently Plan Robust Frictional Multi-Object Grasps](https://mlsanity.com/paper/2210.07420/) (2022)
- [The Busboy Problem: Efficient Tableware Decluttering Using Consolidation and Multi-Object Grasps](https://mlsanity.com/paper/2307.03882/) (2023)
- [Target-mass Grasping of Entangled Food using Pre-grasping & Post-grasping](https://mlsanity.com/paper/2201.00933/) (2022)
- [Multiple-object Grasping Using a Multiple-suction-cup Vacuum Gripper in Cluttered Scenes](https://mlsanity.com/paper/2304.10693/) (2023)
- [Push-MOG: Efficient Pushing to Consolidate Polygonal Objects for Multi-Object Grasping](https://mlsanity.com/paper/2306.14021/) (2023)
- [Only Pick Once -- Multi-Object Picking Algorithms for Picking Exact Number of Objects Efficiently](https://mlsanity.com/paper/2307.02662/) (2023)

## Abstract (arXiv)

Simultaneously grasping and delivering multiple objects can significantly enhance robotic work efficiency and has been a key research focus for decades. The primary challenge lies in determining how to push objects, group them, and execute simultaneous grasping for respective groups while considering object distribution and the hardware constraints of the robot. Traditional rule-based methods struggle to flexibly adapt to diverse scenarios. To address this challenge, this paper proposes an imitation learning-based approach. We collect a series of expert demonstrations through teleoperation and train a diffusion policy network, enabling the robot to dynamically generate action sequences for pushing, grouping, and grasping, thereby facilitating efficient multi-object grasping and delivery. We conducted experiments to evaluate the method under different training dataset sizes, varying object quantities, and real-world object scenarios. The results demonstrate that the proposed approach can effectively and adaptively generate multi-object grouping and grasping strategies. With the support of more training data, imitation learning is expected to be an effective approach for solving the multi-object grasping problem.