Dunli Hu

Papers

Total Citations

H-Index

About

Dunli Hu is a researcher focused on advancing artificial intelligence and human–computer interaction, with key contributions in gesture recognition and autonomous robotic navigation. Hu’s most cited work, “Gesture recognition based on modified Yolov5s” (2022, 13 citations), addresses the growing demand for intuitive human–machine cooperation by optimizing the YOLOv5s deep learning architecture for real-time hand gesture detection. This method enhances accuracy and speed, making it suitable for applications in smart devices and collaborative robotics. In parallel, Hu’s research on reinforcement learning explores the challenge of sparse external rewards in mobile robot navigation. Their paper “A Curiosity-Based Autonomous Navigation Algorithm for Maze Robot” (2022, 2 citations) introduces an intrinsic curiosity mechanism that enables robots to explore and learn effectively even when environmental rewards are minimal. This work contributes to more adaptive and self-directed robotic systems. By combining robust computer vision techniques with innovative learning algorithms, Hu is helping to bridge the gap between human intent and machine action. Their research holds promise for safer, more responsive autonomous systems in manufacturing, service robotics, and assistive technologies.