# FLEX: A Framework for Learning Robot-Agnostic Force-based Skills Involving Sustained Contact Object Manipulation

arXiv: https://arxiv.org/abs/2503.13418 · HTML: https://mlsanity.com/paper/2503.13418/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 30 references that are in the map.
- First posted on arXiv 17 Mar 2025.

## Builds on (most cited first, 20 of 20)

- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation](https://mlsanity.com/paper/1806.10293/) (2018)
- [SAPIEN: A SimulAted Part-based Interactive ENvironment](https://mlsanity.com/paper/2003.08515/) (2020)
- [robosuite: A Modular Simulation Framework and Benchmark for Robot Learning](https://mlsanity.com/paper/2009.12293/) (2020)
- [Deep Reinforcement Learning for Robotic Manipulation with Asynchronous Off-Policy Updates](https://mlsanity.com/paper/1610.00633/) (2016)
- [Accelerating Reinforcement Learning with Learned Skill Priors](https://mlsanity.com/paper/2010.11944/) (2020)
- [Where2Act: From Pixels to Actions for Articulated 3D Objects](https://mlsanity.com/paper/2101.02692/) (2021)
- [A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms](https://mlsanity.com/paper/1907.03146/) (2019)
- [FlowBot3D: Learning 3D Articulation Flow to Manipulate Articulated Objects](https://mlsanity.com/paper/2205.04382/) (2022)
- [VAT-Mart: Learning Visual Action Trajectory Proposals for Manipulating 3D ARTiculated Objects](https://mlsanity.com/paper/2106.14440/) (2021)
- [Augmenting Reinforcement Learning with Behavior Primitives for Diverse Manipulation Tasks](https://mlsanity.com/paper/2110.03655/) (2021)
- [UMPNet: Universal Manipulation Policy Network for Articulated Objects](https://mlsanity.com/paper/2109.05668/) (2021)
- [End-to-End Affordance Learning for Robotic Manipulation](https://mlsanity.com/paper/2209.12941/) (2022)
- [A Survey of Robot Manipulation in Contact](https://mlsanity.com/paper/2112.01942/) (2021)
- [Guided Uncertainty-Aware Policy Optimization: Combining Learning and Model-Based Strategies for Sample-Efficient Policy Learning](https://mlsanity.com/paper/2005.10872/) (2020)
- [Ag2Manip: Learning Novel Manipulation Skills with Agent-Agnostic Visual and Action Representations](https://mlsanity.com/paper/2404.17521/) (2024)
- [Part-Guided 3D RL for Sim2Real Articulated Object Manipulation](https://mlsanity.com/paper/2404.17302/) (2024)
- [AO-Grasp: Articulated Object Grasp Generation](https://mlsanity.com/paper/2310.15928/) (2023)
- [RPMArt: Towards Robust Perception and Manipulation for Articulated Objects](https://mlsanity.com/paper/2403.16023/) (2024)
- [Decoupling Skill Learning from Robotic Control for Generalizable Object Manipulation](https://mlsanity.com/paper/2303.04016/) (2023)

## Built on by (most cited first, 2 of 2)

- [PHIRL: Aligning Learned Rewards with Task Progress for Inverse Reinforcement Learning](https://mlsanity.com/paper/2609.31855/) (2026)
- [Beyond In-Distribution Preservation: Recovering Generalization in Quantized VLAs via Vulnerability-Oriented Tuning](https://mlsanity.com/paper/2610.05745/) (2026)

## Abstract (arXiv)

Learning to manipulate objects efficiently, particularly those involving sustained contact (e.g., pushing, sliding) and articulated parts (e.g., drawers, doors), presents significant challenges. Traditional methods, such as robot-centric reinforcement learning (RL), imitation learning, and hybrid techniques, require massive training and often struggle to generalize across different objects and robot platforms. We propose a novel framework for learning object-centric manipulation policies in force space, decoupling the robot from the object. By directly applying forces to selected regions of the object, our method simplifies the action space, reduces unnecessary exploration, and decreases simulation overhead. This approach, trained in simulation on a small set of representative objects, captures object dynamics -- such as joint configurations -- allowing policies to generalize effectively to new, unseen objects. Decoupling these policies from robot-specific dynamics enables direct transfer to different robotic platforms (e.g., Kinova, Panda, UR5) without retraining. Our evaluations demonstrate that the method significantly outperforms baselines, achieving over an order of magnitude improvement in training efficiency compared to other state-of-the-art methods. Additionally, operating in force space enhances policy transferability across diverse robot platforms and object types. We further showcase the applicability of our method in a real-world robotic setting. For supplementary materials and videos, please visit: https://tufts-ai-robotics-group.github.io/FLEX/