# Reinforcement learning in linear embedding space unlocks generalizable control across soft robot configurations

arXiv: https://arxiv.org/abs/2606.08104 · HTML: https://mlsanity.com/paper/2606.08104/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 16 of its 58 references that are in the map.
- First posted on arXiv 6 Jun 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, 16 of 16)

- [Embodied Intelligence via Learning and Evolution](https://mlsanity.com/paper/2102.02202/) (2021)
- [Linear predictors for nonlinear dynamical systems: Koopman operator meets model predictive control](https://mlsanity.com/paper/1611.03537/) (2016)
- [On Convergence of Extended Dynamic Mode Decomposition to the Koopman Operator](https://mlsanity.com/paper/1703.04680/) (2017)
- [Generalizing Koopman Theory to allow for inputs and control](https://mlsanity.com/paper/1602.07647/) (2016)
- [Lifelike Agility and Play in Quadrupedal Robots using Reinforcement Learning and Generative Pre-trained Models](https://mlsanity.com/paper/2308.15143/) (2023)
- [Active Learning of Dynamics for Data-Driven Control Using Koopman Operators](https://mlsanity.com/paper/1906.05194/) (2019)
- [Soft Robots Modeling: a Structured Overview](https://mlsanity.com/paper/2112.03645/) (2021)
- [Elastica: A compliant mechanics environment for soft robotic control](https://mlsanity.com/paper/2009.08422/) (2020)
- [Derivative-Based Koopman Operators for Real-Time Control of Robotic Systems](https://mlsanity.com/paper/2010.05778/) (2020)
- [Learning Stable Models for Prediction and Control](https://mlsanity.com/paper/2005.04291/) (2020)
- [Robust Tube-based Model Predictive Control with Koopman Operators--Extended Version](https://mlsanity.com/paper/2108.13011/) (2021)
- [Discrete Cosserat Approach for Multi-Section Soft Robots Dynamics](https://mlsanity.com/paper/1702.03660/) (2017)
- [SpiRobs: Logarithmic Spiral-shaped Robots for Versatile Grasping Across Scales](https://mlsanity.com/paper/2303.09861/) (2023)
- [A soft robot that adapts to environments through shape change](https://mlsanity.com/paper/2008.06397/) (2020)
- [Koopman-based Control of a Soft Continuum Manipulator Under Variable Loading Conditions](https://mlsanity.com/paper/2002.01407/) (2020)
- [Topology and morphology design of spherically reconfigurable homogeneous Modular Soft Robots (MSoRos)](https://mlsanity.com/paper/2205.00544/) (2022)

## Abstract (arXiv)

Soft-bodied organisms such as octopuses and elephant trunks exhibit remarkable morphological adaptability, dynamically reconfiguring body shape and stiffness, and flexibly adjusting their control strategies to enable versatile behaviors. Inspired by these biological systems, various soft robots have emerged in recent decades, featuring diverse materials, stiffnesses, and morphologies tailored to specific tasks. Despite substantial advances in the materials and structural designs of soft robots, developing a generalizable control framework capable of rapid adaptation across diverse configurations remains a long-standing challenge. Existing controllers are limited to fixed configurations, demanding laborious configuration-specific remodelling and policy redesign for new configurations. Here, we introduce a generalizable control system that enables rapid adaptation across diverse soft robot configurations via reinforcement learning in a shared linear Koopman embedding space. By encoding robot dynamics into this embedding space, our method decouples control policies from specific morphologies, allowing real-time, model-free policy adaptation across diverse configurations without retraining from scratch. We validate our system across 33 distinct robot configurations. Our system achieves a 75 times reduction in transfer samples across configurations, while sustaining robust performance under high-speed motion, heavy payloads, and multiactuator faults, and achieving real-world skills previously unattainable in soft robotics. This work establishes a unified and adaptable control paradigm for diverse soft robot configurations, bridging mechanical reconfigurability with control flexibility, and may offer broader insights for generalizable control in complex physical systems.