# Pre- and post-contact policy decomposition for non-prehensile manipulation with zero-shot sim-to-real transfer

arXiv: https://arxiv.org/abs/2309.02754 · HTML: https://mlsanity.com/paper/2309.02754/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Sep 2023.

## Built on by (most cited first, 12 of 12)

- [RDT-1B: a Diffusion Foundation Model for Bimanual Manipulation](https://mlsanity.com/paper/2410.07864/) (2024)
- [iVideoGPT: Interactive VideoGPTs are Scalable World Models](https://mlsanity.com/paper/2405.15223/) (2024)
- [TRANSIC: Sim-to-Real Policy Transfer by Learning from Online Correction](https://mlsanity.com/paper/2405.10315/) (2024)
- [DyWA: Dynamics-adaptive World Action Model for Generalizable Non-prehensile Manipulation](https://mlsanity.com/paper/2503.16806/) (2025)
- [XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations](https://mlsanity.com/paper/2511.02776/) (2025)
- [Exploring Embodied Multimodal Large Models: Development, Datasets, and Future Directions](https://mlsanity.com/paper/2502.15336/) (2025)
- [OXE-AugE: A Large-Scale Robot Augmentation of OXE for Scaling Cross-Embodiment Policy Learning](https://mlsanity.com/paper/2512.13100/) (2025)
- [CORN: Contact-based Object Representation for Nonprehensile Manipulation of General Unseen Objects](https://mlsanity.com/paper/2403.10760/) (2024)
- [Learning Visuotactile Estimation and Control for Non-prehensile Manipulation under Occlusions](https://mlsanity.com/paper/2412.13157/) (2024)
- [Learning Extrinsic Dexterity with Parameterized Manipulation Primitives](https://mlsanity.com/paper/2310.17785/) (2023)
- [Learning Long-Horizon Robot Manipulation Skills via Privileged Action](https://mlsanity.com/paper/2502.15442/) (2025)
- [Emerging Extrinsic Dexterity in Cluttered Scenes via Dynamics-aware Policy Learning](https://mlsanity.com/paper/2603.09882/) (2026)

## Abstract (arXiv)

We present a system for non-prehensile manipulation that require a significant number of contact mode transitions and the use of environmental contacts to successfully manipulate an object to a target location. Our method is based on deep reinforcement learning which, unlike state-of-the-art planning algorithms, does not require apriori knowledge of the physical parameters of the object or environment such as friction coefficients or centers of mass. The planning time is reduced to the simple feed-forward prediction time on a neural network. We propose a computational structure, action space design, and curriculum learning scheme that facilitates efficient exploration and sim-to-real transfer. In challenging real-world non-prehensile manipulation tasks, we show that our method can generalize over different objects, and succeed even for novel objects not seen during training. Project website: https://sites.google.com/view/nonprenehsile-decomposition