Аннотация
This paper offers an in-depth look into the design and implementation of a Model Predictive Control (MPC) algorithm for the QUANSER QUBE-SERVO system. The QUBE-SERVO is a sophisticated laboratory experimental setup consisting of a servo motor, an encoder, and a rotary module. This combination provides a robust platform for investigating and testing various control strategies. In particular, the central focus of this study is the usage of the MPC algorithm for controlling the position of the QUBE-SERVO’s rotary disc load module. The MPC algorithm plays a pivotal role in this application by predicting the future behaviors of the system, and controlling the system by minimizing an objective function over a defined finite horizon. This makes it a versatile and effective tool for controlling complex systems. One of the key challenges in practical control applications is maintaining system stability in the presence of disturbances and uncertainties. To this end, we propose a MPC algorithm designed specifically to stabilize the QUBE-SERVO under such conditions. The functionality of this algorithm is not limited to the QUBE-SERVO system alone, and can be extended to other control systems exhibiting similar characteristics. The effectiveness of the proposed MPC algorithm is rigorously tested through simulation studies. These studies involve subjecting the QUBE-SERVO to various reference signals and disturbances. The results of the simulations provide strong evidence of the algorithm’s capability to effectively track reference signals, while also rejecting disturbances and uncertainties, thereby corroborating its efficacy for the QUBE-SERVO application. Moreover, the original MPC algorithm was enhanced to improve its performance for trajectory tracking tasks. We also discuss the integration of the MPC algorithm within the MatLAB and LabVIEW programming environments, which served as the base platforms for designing and running the simulations in this project. This paper, therefore, presents a comprehensive and practical approach for the successful implementation of the MPC algorithm in the QUANSER QUBE-SERVO system, and demonstrates its potential for wider application in similar control systems.