# SLowRL: Safe Low-Rank Adaptation Reinforcement Learning for Locomotion

arXiv: https://arxiv.org/abs/2603.17092 · HTML: https://mlsanity.com/paper/2603.17092/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 19 of its 24 references that are in the map.
- First posted on arXiv 17 Mar 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 19 of 19)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real World](https://mlsanity.com/paper/1703.06907/) (2017)
- [Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey](https://mlsanity.com/paper/2403.14608/) (2024)
- [Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learning](https://mlsanity.com/paper/2109.11978/) (2021)
- [RMA: Rapid Motor Adaptation for Legged Robots](https://mlsanity.com/paper/2107.04034/) (2021)
- [Learning agile and dynamic motor skills for legged robots](https://mlsanity.com/paper/1901.08652/) (2019)
- [Learning robust perceptive locomotion for quadrupedal robots in the wild](https://mlsanity.com/paper/2201.08117/) (2022)
- [Sim-to-Real: Learning Agile Locomotion For Quadruped Robots](https://mlsanity.com/paper/1804.10332/) (2018)
- [Learning Agile Robotic Locomotion Skills by Imitating Animals](https://mlsanity.com/paper/2004.00784/) (2020)
- [Closing the Sim-to-Real Loop: Adapting Simulation Randomization with Real World Experience](https://mlsanity.com/paper/1810.05687/) (2018)
- [Recovery RL: Safe Reinforcement Learning with Learned Recovery Zones](https://mlsanity.com/paper/2010.15920/) (2020)
- [Legged Robots that Keep on Learning: Fine-Tuning Locomotion Policies in the Real World](https://mlsanity.com/paper/2110.05457/) (2021)
- [Constrained Decision Transformer for Offline Safe Reinforcement Learning](https://mlsanity.com/paper/2302.07351/) (2023)
- [Safe Reinforcement Learning for Legged Locomotion](https://mlsanity.com/paper/2203.02638/) (2022)
- [Grow Your Limits: Continuous Improvement with Real-World RL for Robotic Locomotion](https://mlsanity.com/paper/2310.17634/) (2023)
- [Learning-based legged locomotion; state of the art and future perspectives](https://mlsanity.com/paper/2406.01152/) (2024)
- [Model-free Reinforcement Learning for Robust Locomotion using Demonstrations from Trajectory Optimization](https://mlsanity.com/paper/2107.06629/) (2021)
- [Efficient Hyperparameter Optimization for Physics-based Character Animation](https://mlsanity.com/paper/2104.12365/) (2021)
- [Learning to Act Through Contact: A Unified View of Multi-Task Robot Learning](https://mlsanity.com/paper/2510.03599/) (2025)

## Built on by (most cited first, 1 of 1)

- [Mind the Phase: Effective Rank and Representation Health in Legged Locomotion](https://mlsanity.com/paper/2609.06958/) (2026)

## Abstract (arXiv)

Sim-to-real transfer of locomotion policies often leads to performance degradation due to the inevitable sim-to-real gap. Naively fine-tuning these policies directly on hardware is problematic, as it poses risks of mechanical failure and suffers from high sample inefficiency. In this paper, we address the challenge of safely and efficiently fine-tuning reinforcement learning (RL) policies for dynamic locomotion tasks. Specifically, we focus on fine-tuning policies learned in simulation directly on hardware, while explicitly enforcing safety constraints. In doing so, we introduce SLowRL, a framework that combines Low-Rank Adaptation (LoRA) with training-time safety enforcement via a recovery policy. We evaluate our method both in simulation and on a real Unitree Go2 quadruped robot for jump and trot tasks. Experimental results show that our method achieves a 46.5% reduction in fine-tuning time and near-zero safety violations compared to standard proximal policy optimization (PPO) baselines. Notably, we find that a rank-1 adaptation alone is sufficient to recover pre-trained performance in the real world, while maintaining stable and safe real-world fine-tuning. These results demonstrate the practicality of safe, efficient fine-tuning for dynamic real-world robotic applications.