Xiaogang Jia

Chinese People's Armed Police Force, Karlsruhe Institute of Technology

Papers

Total Citations

H-Index

About

Xiaogang Jia’s research bridges the gap between classical control theory and modern robot learning, with a focus on making robots more adaptive and efficient. His early work on adaptive iterative learning control for robot manipulators introduced a self-tuning PD feedback scheme that enables precise trajectory tracking in repetitive tasks, laying a foundation for robust industrial automation. More recently, Jia has pioneered the use of augmented reality for data collection in robot learning, demonstrating in a 2024 study how AR interfaces can streamline the gathering of large-scale demonstrations for training versatile robots. His innovative MaIL architecture, which replaces Transformer-based policies with Mamba state-space models, represents a leap forward in imitation learning, achieving selective attention to key features with improved computational efficiency. With over 20 citations across his most influential papers, Jia’s contributions are shaping the next generation of human-robot interaction and autonomous skill acquisition. His work not only advances theoretical understanding but also provides practical tools for deploying robots in dynamic, real-world environments.