Optimal Control and Sensorless Manipulation of Complex Ensemble Systems

该奖项的研究目标是创建一个统一的控制理论框架，并为复杂集成系统的最优控制和无传感器操作开发计算上可处理的方法。许多实际工程和生物医学应用涉及通过使用一个共同的控制输入来控制结构相似的动力系统的大型集成。构建强大的外部输入，引导系统集合形成所需的动态条件，如同步，是至关重要的。这项研究将产生足够通用的方法，可以应用于跨学科，并可以在从量子物理学到神经科学的各种应用中加以利用。研究方法从基于几何控制技术的集成系统可控性的基础研究发展到开发一种高效的无优化集成控制综合计算方法。交付成果包括用于最优集成控制设计的分布式计算例程，设计控制的实验演示和验证，研究结果的文档，工程学生教育和职前教师的工程研究经验。如果成功，这项研究的结果将大大提高我们对复杂集成系统的理解，并直接促进控制和系统理论的新发展。它还将通过在原子冷却、帕金森病的深部脑刺激和电化学中实现最优控制设计的分析和数值易于处理的方法，从而扩大开创性的应用。这一成果有可能对量子科学技术、网络科学、生物学和神经科学的发展产生重大影响。本科和研究生工程专业的学生和职前教师将受益于课堂教学和参与多学科研究，特别是与实验人员的互动。普通公众和大学预科的K-12学生也将参与其中，例如华盛顿大学的科学拓展项目。

The research objective of this award is to create a unified control-theoretic framework and develop computationally tractable methods for optimal control and sensorless manipulation of complex ensemble systems. Many practical engineering and biomedical applications involve controlling large ensembles of structurally similar dynamical systems by the use of a common control input. Constructing robust external inputs that steer an ensemble of systems to form a desired dynamical condition, such as synchrony, is of fundamental importance. The research will result in methods that are general enough to apply across disciplines and can be exploited in diverse applications ranging from quantum physics to neuroscience. The research approach progresses from a fundamental investigation of controllability of ensemble systems based on geometric control techniques to the development of an efficient optimization-free computational method for optimal ensemble control synthesis. Deliverables include distributed computational routines for optimal ensemble control design, experimental demonstration and validation of the designed controls, documentation of research results, engineering student education, and engineering research experiences for pre-service teachers.If successful, the results of this research will substantially advance our understanding of complex ensemble systems and directly contribute to new developments in control and systems theory. It will also broaden seminal applications by enabling analytically and numerically tractable methods for optimal control designs in atom cooling, deep brain stimulation for Parkinson's disease, and electrochemistry. This achievement has great potential to make significant impacts on the advancement of quantum science and technology, network sciences, biology, and neuroscience. Undergraduate and graduate engineering students and pre-service teachers will benefit through classroom instruction and involvement in multidisciplinary research, in particular gaining interactions with experimentalists. General public and pre-college K-12 students will also be engaged through, for example, Science Outreach Programs at Washington University.