A Study on Trajectory Control System of Hydraulic Excavators Based on Multi-Domain Physical Model*

Abstract

Hydraulic excavators are complex mechatronics construction machinery with characteristics of multidiscipline intersection and multi-domain close coupling. To analyze the comprehensive property of its trajectory control system, a method of multi-system hybrid modeling and simulation based on MATLAB is proposed and described. In this paper, the multi-domain physical model of a medium-sized hydraulic excavator is established. The considered model is mainly comprised of four other subsystems: a machine system, a hydraulic system, a trajectory control system and a sensor system. Moreover, a fuzzy neural network (FNN) PID strategy is introduced to the trajectory control system to guarantee the accuracy of automatic operation. On the basis of the multi-domain physical model, typical simulation experiments for working patterns were performed to validate the performance of the FNNPID controller. Comparison results demonstrate that the precision and velocity response of the FNNPID controller is better than that of the PID with traditional algorithm. The tracking errors of the boom, the arm, the bucket and the swing are decreased by 3∘, 3.2∘, 5.5∘ and 7.5∘, respectively. Establishment of the multi-domain physical model offers technical means for optimization design and rapid modeling of the complex electromechanical system.