# Robot Learning of Shifting Objects for Grasping in Cluttered Environments

arXiv: https://arxiv.org/abs/1907.11035 · HTML: https://mlsanity.com/paper/1907.11035/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Jul 2019.

## Built on by (most cited first, 13 of 13)

- [Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization](https://mlsanity.com/paper/2601.12993/) (2026)
- [SpeedFolding: Learning Efficient Bimanual Folding of Garments](https://mlsanity.com/paper/2208.10552/) (2022)
- [Free-form language-based robotic reasoning and grasping](https://mlsanity.com/paper/2503.13082/) (2025)
- [Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization](https://mlsanity.com/paper/2602.09722/) (2026)
- [A Hybrid Approach for Learning to Shift and Grasp with Elaborate Motion Primitives](https://mlsanity.com/paper/2111.01510/) (2021)
- [Uncertainty-driven Exploration Strategies for Online Grasp Learning](https://mlsanity.com/paper/2309.12038/) (2023)
- [Graph learning in robotics: a survey](https://mlsanity.com/paper/2310.04294/) (2023)
- [DMFC-GraspNet: Differentiable Multi-Fingered Robotic Grasp Generation in Cluttered Scenes](https://mlsanity.com/paper/2308.00456/) (2023)
- [Pseudo-Labeling and Contextual Curriculum Learning for Online Grasp Learning in Robotic Bin Picking](https://mlsanity.com/paper/2403.02495/) (2024)
- [URPlanner: A Universal Paradigm For Collision-Free Robotic Motion Planning Based on Deep Reinforcement Learning](https://mlsanity.com/paper/2505.20175/) (2025)
- [Visual Foresight With a Local Dynamics Model](https://mlsanity.com/paper/2206.14802/) (2022)
- [Knolling Bot: Teaching Robots the Human Notion of Tidiness](https://mlsanity.com/paper/2310.04566/) (2023)
- [Visual Prompt Guided Unified Pushing Policy](https://mlsanity.com/paper/2602.19193/) (2026)

## Abstract (arXiv)

Robotic grasping in cluttered environments is often infeasible due to obstacles preventing possible grasps. Then, pre-grasping manipulation like shifting or pushing an object becomes necessary. We developed an algorithm that can learn, in addition to grasping, to shift objects in such a way that their grasp probability increases. Our research contribution is threefold: First, we present an algorithm for learning the optimal pose of manipulation primitives like clamping or shifting. Second, we learn non-prehensible actions that explicitly increase the grasping probability. Making one skill (shifting) directly dependent on another (grasping) removes the need of sparse rewards, leading to more data-efficient learning. Third, we apply a real-world solution to the industrial task of bin picking, resulting in the ability to empty bins completely. The system is trained in a self-supervised manner with around 25000 grasp and 2500 shift actions. Our robot is able to grasp and file objects with 274 picks per hour. Furthermore, we demonstrate the system's ability to generalize to novel objects.