Adaptive Task-Space Control for Hydraulic Excavators Based on the High-Order Fully Actuated System Approach
提出一种基于高阶全驱系统方法的液压挖掘机控制框架，通过自适应神经网络补偿多源不确定性，实现任务空间高精度跟踪控制，仿真和实验验证了有效性。
This article proposes a novel control framework for hydraulic excavators based on a high-order fully actuated (HOFA) system approach. First, a comprehensive HOFA model is established, which describes the excavator dynamics in task space, joint space, and drive space. The proposed control algorithm systematically addresses multisource uncertainties, including kinematic calibration errors, as well as structured and unstructured parameter uncertainties in the dynamic and actuator models, via physically derived adaptive neural network compensation integrated into the controller. By decoupling the kinematic and dynamic loops, the algorithm simplifies controller design and theoretical analysis. Furthermore, by integrating the HOFA approach with task-space sensory feedback, the controller enables direct task specification and high-precision control of the excavator bucket tip. Finally, Lyapunov-based theoretical analysis proves asymptotic convergence of task-space tracking errors, and both simulation and experimental results validate the effectiveness of the proposed algorithm.