Direct Optimization Based Compensation Adaptive Robust Control of Nonlinear Systems With State and Input Constraints

• Published in: IEEE transactions on industrial informatics Vol. 17; no. 8; pp. 5441 - 5449
• Main Authors: Liu, Xingyi, Chen, Zheng, Liu, Yingqiang, Duan, Fuxin, Yang, Can, Yao, Bin
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.08.2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Adaptation models; Adaptive control; Adaptive robust control (ARC); Compensation; Control theory; Electric motors; Friction; Informatics; input saturation; linear motor; Motion control; Nonlinear control; Nonlinear systems; Nonlinearity; online; Optimization; Robust control; state constraints; Tracking; Uncertainty
• DOI: 10.1109/TII.2020.3032424
• ISSN: 1551-3203

Motion control of mechatronic system with state and input constraints while achieving excellent integrated performance, such as robustness, high tracking accuracy, fast response, and slow overshoot, has always been a challenging issue. However, most existing relating studies merely focus on how to ensure stability under constraints, and few take integrated performance into account. In this article, we proposed a direct optimization based compensation adaptive robust control (ARC) approach, which is under a two-loop feedback structure, where the outer loop directly online replans both the model compensation term and the reference that conform to the constraints; and the conventional ARC control law is synthesized in inner loop to ensure guaranteed tracking accuracy when facing nonlinearity, parametric uncertainties, and external disturbances. Motion control of a linear motor was considered through this article as an introductory example. Comparative experiments are carried out and the results further verify the superiority and effectiveness of the proposed scheme.