A Framework for Soft Robot Control: Integrating Physics-Based Modeling with Exploration Based Learning
Costanza Armanini, Anthony Tzes, Andrea Del Prete, Fares Abu-Dakka
AI summary
Problem
Soft robots are difficult to control due to their nonlinear deformations and infinite degrees of freedom, creating a trade-off between computationally heavy physics-based models and data-hungry, less robust learning-based policies.
Approach
The authors train a Proximal Policy Optimization agent within a physics-based simulation environment using a compact Geometric Variable-Strain formulation, allowing the agent to learn accurate control policies from synthetic, physically consistent data.
Key results
- 100% success rate in a simulated basketball throw task
- 55% of throw trials achieved 1 cm precision
- Maximum 8.5% relative error in a multi-step pick-and-place task
- Real-time control capability without exhaustive real-world data collection
Why it matters
Offers a scalable, robust control framework for soft robotics that bridges the gap between high-fidelity physics modeling and efficient machine learning, benefiting applications in unstructured environments.
Abstract
Soft robots present unique challenges for accurate modeling and control due to their virtually infinite degrees of freedom and highly nonlinear deformations. High-fidelity continuum models offer accuracy but are often computationally prohibitive for real-time control, while purely learning-based policies are efficient yet frequently lack robustness and require extensive data collection. In this paper, we propose a hybrid control framework that trains Reinforcement Learning (RL) policies using the physics-based Geometric Variable-Strain (GVS) formulation. This integration enables a Proximal Policy Optimization (PPO) agent to learn on a compact, physically exact state parameterization within the SoRoSim environment, leveraging continuum mechanics for accuracy without the need for real-world data collection. We validate our approach in simulation through two tasks: a basketball throw task and a multi-step pick-and-place scenario. In the throwing task, the agent achieved a 100% success rate within a defined tolerance, with 55% of trials reaching the target with 1cm precision. In the multi-step scenario, the controller maintained high accuracy with a maximum relative error of approximately 8.5%. These results demonstrate that combining GVS-based modeling with PPO yields robust, data-efficient control policies, providing a scalable solution for controlling soft robotic systems across diverse applications.