Iterative Learning Control Using a Novel Dynamic Quantizer

This paper investigates the challenges associated with quantized learning control in linear networked systems, aiming to enhance the tracking performance and reduce communication burden on the network. To address these challenges, we propose a novel solution that optimizes both the quantization interval and scaling parameters of a dynamic quantizer integrated with a learning control scheme. We demonstrate that within this design framework, the system can achieve zero-error tracking performance without imposing constraints on the saturation bounds of the quantizer. To validate the effectiveness of our proposed approach, we present simulation results that clearly illustrate its performance.