# Learning Stable Normalizing-Flow Control for Robotic Manipulation

arXiv: https://arxiv.org/abs/2011.00072 · HTML: https://mlsanity.com/paper/2011.00072/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Oct 2020.

## Built on by (most cited first, 9 of 9)

- [Riemannian Flow Matching Policy for Robot Motion Learning](https://mlsanity.com/paper/2403.10672/) (2024)
- [Fast and Robust Visuomotor Riemannian Flow Matching Policy](https://mlsanity.com/paper/2412.10855/) (2024)
- [Generative Modelling of Stochastic Actions with Arbitrary Constraints in Reinforcement Learning](https://mlsanity.com/paper/2311.15341/) (2023)
- [Multiscale Sensor Fusion and Continuous Control with Neural CDEs](https://mlsanity.com/paper/2203.08715/) (2022)
- [CEIP: Combining Explicit and Implicit Priors for Reinforcement Learning with Demonstrations](https://mlsanity.com/paper/2210.09496/) (2022)
- [SERNF: Sample-Efficient Real-World Dexterous Policy Fine-Tuning via Action-Chunked Critics and Normalizing Flows](https://mlsanity.com/paper/2602.09580/) (2026)
- [Data-driven identification of nonlinear dynamical systems with LSTM autoencoders and Normalizing Flows](https://mlsanity.com/paper/2503.03977/) (2025)
- [Issues with Measuring Task Complexity via Random Policies in Robotic Tasks](https://mlsanity.com/paper/2602.18856/) (2026)
- [Potential-Field Action Representation for Reinforcement Learning in Contact-Rich Manipulation](https://mlsanity.com/paper/2609.21609/) (2026)

## Abstract (arXiv)

Reinforcement Learning (RL) of robotic manipulation skills, despite its impressive successes, stands to benefit from incorporating domain knowledge from control theory. One of the most important properties that is of interest is control stability. Ideally, one would like to achieve stability guarantees while staying within the framework of state-of-the-art deep RL algorithms. Such a solution does not exist in general, especially one that scales to complex manipulation tasks. We contribute towards closing this gap by introducing textitnormalizing-flow control structure, that can be deployed in any latest deep RL algorithms. While stable exploration is not guaranteed, our method is designed to ultimately produce deterministic controllers with provable stability. In addition to demonstrating our method on challenging contact-rich manipulation tasks, we also show that it is possible to achieve considerable exploration efficiency--reduced state space coverage and actuation efforts--without losing learning efficiency.