Abstract:
For a novel 3-PRRR/PPRR redundantly actuated parallel mechanism, the dynamic modeling and sliding mode control are presented. Firstly, the kinematic analysis of the paral...Show MoreMetadata
Abstract:
For a novel 3-PRRR/PPRR redundantly actuated parallel mechanism, the dynamic modeling and sliding mode control are presented. Firstly, the kinematic analysis of the parallel mechanism is completed and the dynamic model of the parallel mechanism is established by the Lagrange formulation. In order to improve the computation efficiency of the model and thus improve the rapid performance of model-based control system for the parallel mechanism, a simplification strategy for the dynamic modeling of the parallel mechanism is proposed based on its motion characteristics. The errors caused by the model simplification are analyzed. Afterwards, a simplified dynamic model with the intelligent error compensation is established. It is shown from the results of MATLAB simulations on the non-simplified dynamic model, the simplified dynamic model and the simplified dynamic model with the intelligent error compensation of the parallel mechanism that the simplified dynamic model with the neural network error compensation not only has the higher computation efficiency, but also keeps the higher model precision, compared with the non-simplified dynamic model of the parallel mechanism. Then, the sliding mode controllers based on the non-simplified dynamic model, the simplified dynamic model and the simplified dynamic model with the neural network error compensation are respectively designed in a task space. The stability of the controller based on the simplified dynamic model with the neural network error compensation is proved theoretically. Finally, the simulation and experimental results verify the correctness of the simplified dynamic model with the neural network error compensation and the feasibility and effectiveness of the sliding mode controller based on the model.
Date of Conference: 01-03 June 2016
Date Added to IEEE Xplore: 07 July 2016
ISBN Information: