Integrated Direct and Indirect Adaptive Robust Control With Quantitative Robust Parameter Estimation -- Theory and Applications

当今的工业要求机器不仅要生产高质量的产品和高生产率，还要具有内置的智能，如预测和诊断功能，可以减少机器停机时间和维护成本。这些工业发展趋势对面向性能的先进控制提出了更高的要求。此外，控制系统应具有实时学习能力，以帮助构建智能功能。拟议的研究是朝着满足这些工业需求迈出的一步。具体地说，该项目的目标是：（i）开发一个通用的理论框架，用于设计新一代面向性能的非线性自适应鲁棒控制器（ARC），这些控制器是实用的，明确地考虑到过程非线性和不确定性的影响，并具有精确的在线参数估计，用于次要目的，如机器部件健康监测和预测;（ii）将新的集成ARC设计应用于现代机械系统的智能和精确控制，例如高速电磁马达驱动的机械系统。所提出的集成ARC设计不仅具有重要的实用价值，而且通过弥合直接和间接两类不同的自适应控制设计之间的差距，为自适应控制界做出了基本的理论贡献。提出的定量鲁棒参数估计将产生深远的实际影响，因为它也是自动化机载建模等其他工业技术的关键推动因素。拟议的研究包括两个主要领域的控制，已在大多数大学教授：非线性和自适应。建议的应用程序是那些工程专业的学生在他们的职业生涯中最有可能遇到的：液压驱动的重型机器和电机驱动的精密设备。研究结果将对未来的控制课程，在本科和研究生水平产生持久的影响。拟议的研究为新一代控制器的设计提供了坚实的基础，这些控制器将帮助工业界建造具有卓越性能和智能的现代机器。此外，NSF的支持将帮助PI继续目前的工业合作，并建立新的工业和国际合作，这在Maxtor Inc.的支持信中显而易见。

Today's industry demands machines that will not only produce high quality products with high productivity, but also feature built-in intelligence like prognostic and diagnostic capabilities that can reduce machine down-times and maintenance costs. These industrial trends put a great demand on performance oriented advanced controls. Additionally, the control system should possess a real-time learning capability to aid the construction of intelligent features. The proposed research is a step toward meeting these industrial demands. Specifically, the objectives of the project are: (i) to develop a general theoretical framework for the design of a new generation of performance oriented nonlinear adaptive robust controllers (ARC) that are practical, explicitly take into account the effects of process nonlinearities and uncertainties, and have accurate on-line parameter estimation for secondary purposes such as machine component health monitoring and prognosis; (ii) to apply the new integrated ARC design to the intelligent and precision control of modern mechanical systems such as high-speed electro-magnetic motor driven ones. The proposed integrated ARC design will not only have significant practical value, but also make fundamental theoretical contributions to the adaptive control community by bridging the gap between the two distinct classes of adaptive control designs: direct and indirect. The proposed quantitative robust parameter estimation will have a profound practical impact since it is also a key enabler for other industrial technologies such as automated on-board modeling. The proposed research encompasses two major fields of controls that have been taught in most universities: nonlinear and adaptive. The proposed applications are those that engineering students will most likely encounter in their career: hydraulically actuated heavy machines and motor driven precision devices. The research results will have a lasting impact on future control curriculum, at both the undergraduate and the graduate level. The proposed research provides a solid foundation for the design of a new generation of controllers that will help industry build modern machines of great performance and intelligence. In addition, the NSF support will help the PI continue current industrial collaborations and to establish new industrial as well as international ones, as evident in the support letter from Maxtor Inc.