Hangyao Tu

Zhejiang University of Science and Technology, NetEase (China)

Papers

Total Citations

H-Index

About

Hangyao Tu is a robotics researcher whose work focuses on autonomous navigation and 3D perception for intelligent systems. Their key contributions lie in two critical areas: robot localization and path planning, and 6D object pose estimation for robotic manipulation. In their 2024 paper on AGV localization, Tu addressed fundamental limitations in adaptive Monte Carlo localization (AMCL) by introducing a vision-based initial localization method combined with the PO-JPS path planning algorithm, solving problems of global localization failure and computational inefficiency. This work has already garnered 4 citations, reflecting its practical relevance. For robotic manipulation, Tu proposed DON6D, a decoupled one-stage network for 6D pose estimation that overcomes the slow inference speeds of traditional two-stage solutions while handling challenging real-world conditions like occlusion, lighting variation, and sensor noise. This innovation, with 2 citations, represents a significant step toward real-time, robust perception for grasping tasks. Tu’s research bridges the gap between theoretical algorithms and practical deployment, making autonomous systems more reliable and efficient in dynamic environments.