TD3-Optimized LADRC for Motion Smoothness Control of Hydraulic Manipulators

To address the insufficient motion smoothness of a hydraulic manipulator under complex multi-source disturbances, this paper proposes a composite control strategy that performs online scheduling of Linear Active Disturbance Rejection Control (LADRC) parameters based on offline pre-training using Twin Delayed Deep Deterministic Policy Gradient (TD3). First, based on the dynamic model of a joint actuated by a valve-controlled asymmetric cylinder, an improved third-order LADRC with observation error compensation is designed, where a linear extended state observer (LESO) provides real-time estimation and feedforward compensation of uncertainties and disturbances. Then, by analyzing the influence mechanism of the core LADRC parameters on the manipulator motion smoothness, a TD3-based online parameter self-tuning framework is constructed. A multi-objective reward function integrating tracking error, jerk, and control input smoothness is designed to achieve adaptive optimization of observer bandwidth and controller gain. Finally, an electro–mechanical–hydraulic co-simulation platform and a full-scale experimental test rig for a rock-drilling manipulator are established, and the effectiveness of the proposed control strategy is validated through extensive simulations and physical experiments. Simulation results show that, compared with Actor–Critic PID, the maximum trajectory tracking error is reduced from 79.4 mm to within 10.8 mm. Physical experiments demonstrate that compared with PSO‑LADRC, the root‑mean‑square error is reduced by 65.38% and the peak angular jerk is decreased by 55.81%, which effectively improves motion smoothness under complex operating conditions.

Authors

Institutions

Publication Details

Journal
Applied Sciences
Published
2026-11-13
DOI
https://doi.org/10.3390/app162211259
Primary Topic
Hydraulic and Pneumatic Systems
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

TD3-Optimized LADRC for Motion Smoothness Control of Hydraulic Manipulators

Shuqi Lan, Yue Sun, Qiaoyu Xu
2 links
Applied Sciences
Hydraulic and Pneumatic Systems
article

TD3-Optimized LADRC for Motion Smoothness Control of Hydraulic Manipulators

Shuqi Lan, Yue Sun, Qiaoyu Xu
article en

Abstract

To address the insufficient motion smoothness of a hydraulic manipulator under complex multi-source disturbances, this paper proposes a composite control strategy that performs online scheduling of Linear Active Disturbance Rejection Control (LADRC) parameters based on offline pre-training using Twin Delayed Deep Deterministic Policy Gradient (TD3). First, based on the dynamic model of a joint actuated by a valve-controlled asymmetric cylinder, an improved third-order LADRC with observation error compensation is designed, where a linear extended state observer (LESO) provides real-time estimation and feedforward compensation of uncertainties and disturbances. Then, by analyzing the influence mechanism of the core LADRC parameters on the manipulator motion smoothness, a TD3-based online parameter self-tuning framework is constructed. A multi-objective reward function integrating tracking error, jerk, and control input smoothness is designed to achieve adaptive optimization of observer bandwidth and controller gain. Finally, an electro–mechanical–hydraulic co-simulation platform and a full-scale experimental test rig for a rock-drilling manipulator are established, and the effectiveness of the proposed control strategy is validated through extensive simulations and physical experiments. Simulation results show that, compared with Actor–Critic PID, the maximum trajectory tracking error is reduced from 79.4 mm to within 10.8 mm. Physical experiments demonstrate that compared with PSO‑LADRC, the root‑mean‑square error is reduced by 65.38% and the peak angular jerk is decreased by 55.81%, which effectively improves motion smoothness under complex operating conditions.

Applied SciencesVol. 16(22)
Henan University of Science and Technology (CN)
Hydraulic and Pneumatic Systems
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon