Modified reduced-order modeling of a flexible robot-manipulator and model-associative vibration control implementation

Abstract

This paper introduces a modified reduced-order model with aim to use in the dynamic analysis of flexible robot-manipulators. The model structure is firstly discussed theoretically, then numerically, and finally experimentally in terms of efficiency, utility and robustness. Due to the deterministic structure of the model, which is formed by taking into account the specific features of the system, a successful model can be created without the need for wide range of training data such in probabilistic methods. Because of the simplicity of the model, it is able to realize a dramatic decrease in the analysis period compared to FEA programs. The model is also quite successful in reflecting the robot's nonlinear trigonometric position responses. Besides, an open loop model-associative vibration control method was used to show the utility and satisfactory results were achieved in the actual system. Furthermore, the ability of the model to perform multiple scenarios in a short time allowed the case-study of the control input parameters. © 2020 Elsevier Ltd