Reinforcement Learning Controllers for Soft Robots using Learned Environments
On the map: Deep reinforcement learning methods → RL for robot locomotion and manipulation
Abstract (arXiv)
Soft robotic manipulators offer operational advantage due to their compliant and deformable structures. However, their inherently nonlinear dynamics presents substantial challenges. Traditional analytical methods often depend on simplifying assumptions, while learning-based techniques can be computationally demanding and limit the control policies to existing data. This paper introduces a novel approach to soft robotic control, leveraging state-of-the-art policy gradient methods within parallelizable synthetic environments learned from data. We also propose a safety oriented actuation space exploration protocol via cascaded updates and weighted randomness. Specifically, our recurrent forward dynamics model is learned by generating a training dataset from a physically safe textitmean reverting random walk in actuation space to explore the partially-observed state-space. We demonstrate a reinforcement learning approach towards closed-loop control through state-of-the-art actor-critic methods, which efficiently learn high-performance behaviour over long horizons. This approach removes the need for any knowledge regarding the robot's operation or capabilities and sets the stage for a comprehensive benchmarking tool in soft robotics control.
Key facts. Cited by 3 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 6 of its 27 references that are in the map. First posted on arXiv 24 Oct 2024.
Builds on
6 of 27 references.
References · 6
- 2017Proximal Policy Optimization Algorithmscited by 15,215
- 2015Continuous control with deep reinforcement learningcited by 3,407
- 2022Robot Operating System 2: Design, Architecture, and Uses In The Wildcited by 199
- 2022Discovered Policy Optimisationcited by 104
- 2021Podracer architectures for scalable Reinforcement Learningcited by 38
- 2017Analyzing and Exploiting NARX Recurrent Neural Networks for Long-Term Dependenciescited by 2
Built on by
Citing papers per year: 2025: 3.
3 papers.
- 2025Hysteresis-Aware Neural Network Modeling and Whole-Body Reinforcement Learning Control of Soft Robotscited by 1
- 2025Contact-Aided Navigation of Flexible Robotic Endoscope Using Deep Reinforcement Learning in Dynamic Stomachcited by 0
- 2025DiSA-IQL: Offline Reinforcement Learning for Robust Soft Robot Control under Distribution Shiftscited by 0
- 2025DiSA-IQL: Offline Reinforcement Learning for Robust Soft Robot Control under Distribution Shifts30 Sep 2025
- 2025Contact-Aided Navigation of Flexible Robotic Endoscope Using Deep Reinforcement Learning in Dynamic Stomach30 Aug 2025
- 2025Hysteresis-Aware Neural Network Modeling and Whole-Body Reinforcement Learning Control of Soft Robots18 Apr 2025
Similar, not cited
Close in wording, no citation either way
- 2025Real-Time Reinforcement Learning for Dynamic Tasks with a Parallel Soft Robot
- 2025Learning Soft Robotic Dynamics with Active Exploration
- 2025Sensor-Space Based Robust Kinematic Control of Redundant Soft Manipulator by Learning
- 2025SoftMimic: Learning Compliant Whole-body Control from Examples
- 2025Deep Reinforcement Learning-Based Motion Planning and PDE Control for Flexible Manipulators
- 2026Reinforcement learning in linear embedding space unlocks generalizable control across soft robot configurations
- 2025Learning to Crawl: Latent Model-Based Reinforcement Learning for Soft Robotic Adaptive Locomotion
- 2026Physics-Informed Policy Optimization via Analytic Dynamics Regularization