About the project
This PhD project aims to advance iterative learning control (ILC) by eliminating the dependency on analytical models, which are often costly or impractical to obtain. By leveraging data-driven control and optimization methods, this research will develop novel ILC algorithms that achieve high convergence performance directly from data.
Iterative Learning Control (ILC) has demonstrated impressive convergence capabilities, traditionally relying on precise analytical models to achieve this performance. However, obtaining these models can be challenging, time-consuming, and costly in many practical applications, making model-based ILC approaches less feasible.
This PhD project seeks to address this limitation by developing innovative ILC algorithms that achieve rapid convergence directly from data, eliminating the need for explicit system models.
The project will leverage cutting-edge tools from data-driven control and optimization to design algorithms that can learn and refine control actions based solely on data obtained from system trials. Data-driven approaches allow systems to adapt their behavior without the limitations and costs associated with model identification, making these methods particularly suited to complex, variable, or uncertain environments where system modeling is difficult or impractical. By focusing on optimizing convergence performance from data alone, the research will contribute to a new generation of ILC techniques that are adaptable, cost-effective, and highly scalable.
The successful candidate will develop and test new data-driven ILC frameworks, evaluating their effectiveness in simulated and real-world environments. These advancements are expected to significantly enhance the applicability of ILC across various fields, including robotics, autonomous systems, and manufacturing, where high-performance learning without costly model identification is increasingly in demand.
The candidate will gain experience at the intersection of data-driven control, optimization, and iterative learning, positioning them at the forefront of this dynamic research area.