# Policy Transfer via Kinematic Domain Randomization and Adaptation

arXiv: https://arxiv.org/abs/2011.01891 · HTML: https://mlsanity.com/paper/2011.01891/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 3 Nov 2020.

## Built on by (most cited first, 16 of 16)

- [MetaMorph: Learning Universal Controllers with Transformers](https://mlsanity.com/paper/2203.11931/) (2022)
- [RoboGSim: A Real2Sim2Real Robotic Gaussian Splatting Simulator](https://mlsanity.com/paper/2411.11839/) (2024)
- [Transform2Act: Learning a Transform-and-Control Policy for Efficient Agent Design](https://mlsanity.com/paper/2110.03659/) (2021)
- [Robust High-speed Running for Quadruped Robots via Deep Reinforcement Learning](https://mlsanity.com/paper/2103.06484/) (2021)
- [From Screens to Scenes: A Survey of Embodied AI in Healthcare](https://mlsanity.com/paper/2501.07468/) (2025)
- [High-Performance Reinforcement Learning on Spot: Optimizing Simulation Parameters with Distributional Measures](https://mlsanity.com/paper/2504.17857/) (2025)
- [Robust 3D Shape Reconstruction in Zero-Shot from a Single Image in the Wild](https://mlsanity.com/paper/2403.14539/) (2024)
- [Curriculum Reinforcement Learning via Morphology-Environment Co-Evolution](https://mlsanity.com/paper/2309.12529/) (2023)
- [Phys2Real: Fusing VLM Priors with Interactive Online Adaptation for Uncertainty-Aware Sim-to-Real Manipulation](https://mlsanity.com/paper/2510.11689/) (2025)
- [Towards Sample-Efficiency and Generalization of Transfer and Inverse Reinforcement Learning: A Comprehensive Literature Review](https://mlsanity.com/paper/2411.10268/) (2024)
- [Towards Real-World Efficiency: Domain Randomization in Reinforcement Learning for Pre-Capture of Free-Floating Moving Targets by Autonomous Robots](https://mlsanity.com/paper/2406.06460/) (2024)
- [Characterization of Human Balance through a Reinforcement Learning-based Muscle Controller](https://mlsanity.com/paper/2308.04462/) (2023)
- [Error-Aware Policy Learning: Zero-Shot Generalization in Partially Observable Dynamic Environments](https://mlsanity.com/paper/2103.07732/) (2021)
- [Unified Simulation, Perception, and Generation of Human Behavior](https://mlsanity.com/paper/2204.13678/) (2022)
- [RoboPearls: Editable Video Simulation for Robot Manipulation](https://mlsanity.com/paper/2506.22756/) (2025)
- [Sim-to-Real of Humanoid Locomotion Policies via Joint Torque Space Perturbation Injection](https://mlsanity.com/paper/2603.21853/) (2026)

## Abstract (arXiv)

Transferring reinforcement learning policies trained in physics simulation to the real hardware remains a challenge, known as the "sim-to-real" gap. Domain randomization is a simple yet effective technique to address dynamics discrepancies across source and target domains, but its success generally depends on heuristics and trial-and-error. In this work we investigate the impact of randomized parameter selection on policy transferability across different types of domain discrepancies. Contrary to common practice in which kinematic parameters are carefully measured while dynamic parameters are randomized, we found that virtually randomizing kinematic parameters (e.g., link lengths) during training in simulation generally outperforms dynamic randomization. Based on this finding, we introduce a new domain adaptation algorithm that utilizes simulated kinematic parameters variation. Our algorithm, Multi-Policy Bayesian Optimization, trains an ensemble of universal policies conditioned on virtual kinematic parameters and efficiently adapts to the target environment using a limited number of target domain rollouts. We showcase our findings on a simulated quadruped robot in five different target environments covering different aspects of domain discrepancies.