Contraction Metric Based Safe Reinforcement Learning Force Control for a Hydraulic Actuator with Real-World Training

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Maitan, Lucca, Toschi, Lucas, Zanette, Cícero, Vergamini, Elisa G., Santos, Leonardo F., Boaventura, Thiago
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918328857001984
author Maitan, Lucca
Toschi, Lucas
Zanette, Cícero
Vergamini, Elisa G.
Santos, Leonardo F.
Boaventura, Thiago
author_facet Maitan, Lucca
Toschi, Lucas
Zanette, Cícero
Vergamini, Elisa G.
Santos, Leonardo F.
Boaventura, Thiago
contents Force control in hydraulic actuators is notoriously difficult due to strong nonlinearities, uncertainties, and the high risks associated with unsafe exploration during learning. This paper investigates safe reinforcement learning (RL) for hy draulic force control with real-world training using contraction metric certificates. A data-driven model of a hydraulic actuator, identified from experimental data, is employed for simulation based pretraining of a Soft Actor-Critic (SAC) policy that adapts the PI gains of a feedback-linearization (FL) controller. To reduce instability during online training, we propose a quadratic-programming (QP) contraction filter that leverages a learned contraction metric to enforce approximate exponential convergence of trajectories, applying minimal corrections to the policy output. The approach is validated on a hydraulic test bench, where the RL controller is trained directly on hardware and benchmarked against a simulation-trained agent and a fixed-gain baseline. Experimental results show that real-hardware training improves force-tracking performance compared to both alternatives, while the contraction filter mitigates chattering and instabilities. These findings suggest that contraction-based certificates can enable safe RL in high force hydraulic systems, though robustness at extreme operating conditions remains a challenge.
format Preprint
id arxiv_https___arxiv_org_abs_2602_08977
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Contraction Metric Based Safe Reinforcement Learning Force Control for a Hydraulic Actuator with Real-World Training
Maitan, Lucca
Toschi, Lucas
Zanette, Cícero
Vergamini, Elisa G.
Santos, Leonardo F.
Boaventura, Thiago
Systems and Control
Force control in hydraulic actuators is notoriously difficult due to strong nonlinearities, uncertainties, and the high risks associated with unsafe exploration during learning. This paper investigates safe reinforcement learning (RL) for hy draulic force control with real-world training using contraction metric certificates. A data-driven model of a hydraulic actuator, identified from experimental data, is employed for simulation based pretraining of a Soft Actor-Critic (SAC) policy that adapts the PI gains of a feedback-linearization (FL) controller. To reduce instability during online training, we propose a quadratic-programming (QP) contraction filter that leverages a learned contraction metric to enforce approximate exponential convergence of trajectories, applying minimal corrections to the policy output. The approach is validated on a hydraulic test bench, where the RL controller is trained directly on hardware and benchmarked against a simulation-trained agent and a fixed-gain baseline. Experimental results show that real-hardware training improves force-tracking performance compared to both alternatives, while the contraction filter mitigates chattering and instabilities. These findings suggest that contraction-based certificates can enable safe RL in high force hydraulic systems, though robustness at extreme operating conditions remains a challenge.
title Contraction Metric Based Safe Reinforcement Learning Force Control for a Hydraulic Actuator with Real-World Training
topic Systems and Control
url https://arxiv.org/abs/2602.08977