Online Multi‐Constraint Estimation and Correction for Path Tracking of Hydraulic Robotic Manipulators

ABSTRACT Precise tool‐center‐point (TCP) path tracking is essential for hydraulic robotic manipulators (HRMs) in tasks requiring high accuracy and efficiency. Achieving optimal performance requires minimizing execution time while maintaining superior path tracking. However, inherent system uncertainties and time‐varying operational constraints often challenge the effectiveness of existing methods. These methods may either violate critical constraints—compromising tracking accuracy—or fail to meet the real‐time demands required for online applications. To overcome these limitations, this paper presents a novel path tracking method specifically designed for TCP path tracking in HRMs. The proposed method formulates specialized, time‐varying, multi‐constraint estimation models tailored to the characteristics of HRMs. Within this framework, trajectory planning is decoupled from control execution, allowing dynamic correction of time‐varying constraints based on the TCP's tracking state. This strategy effectively compensates for system uncertainties. Furthermore, an efficient nonlinear filter (NF) is incorporated to ensure real‐time trajectory planning performance. The effectiveness of the proposed method is validated on an HRM through comparative experiments involving two path types, demonstrating improved path tracking accuracy.