Xiaofan Yu

Papers

Total Citations

H-Index

About

Xiaofan Yu is a researcher at the forefront of intelligent robotics, with a primary focus on integrating deep reinforcement learning into real-world autonomous systems. Her most-cited work, "A Robotic Auto-Focus System based on Deep Reinforcement Learning" (2018, 12 citations), pioneers an end-to-end approach that leverages Deep Q Networks (DQN) to replace traditional auto-focus methods. By demonstrating how reinforcement learning can handle high-dimensional visual inputs and learn discrete control policies, Yu’s contribution directly addresses a critical bottleneck in robotic vision—enabling cameras to autonomously and adaptively focus without manual calibration. This work not only showcases her expertise in merging computer vision with decision-making algorithms but also lays the groundwork for more responsive, learning-driven robotic systems. While her citation count reflects the niche yet foundational nature of her research, the impact is clear: her methodology offers a scalable alternative to conventional control, inspiring further exploration into reinforcement learning for low-level robotic tasks. Yu’s research is particularly notable for its practical orientation, bridging theoretical reinforcement learning with tangible hardware applications. For students and researchers, her work exemplifies how cutting-edge AI techniques can solve persistent engineering challenges, making her a compelling figure in the evolution of autonomous robotics.