Haoxuan Song
Papers
5
Total Citations
51
H-Index
4
About
Haoxuan Song is a researcher at the forefront of sustainable manufacturing, robotics, and computer vision, whose work bridges the gap between intelligent automation and environmental responsibility. His primary research areas include disassembly line balancing, human-robot collaboration, and 3D scene reconstruction. Song’s most impactful contribution is his pioneering work on optimizing stochastic disassembly systems, where he developed a hybrid evolutionary algorithm to minimize carbon emissions while balancing complex, uncertain workflows—a paper that has already garnered 21 citations since 2024. He further advanced the field by modeling and optimizing line efficiency for preventive maintenance in robotic disassembly, earning 15 citations in 2025. In computer vision, Song introduced HDR-Net-Fusion, a hierarchical deep reinforcement network enabling real-time 3D dynamic scene reconstruction, and proposed a novel method for joint hand and object pose estimation from single RGB images using high-level 2D constraints. His work also extends to software reliability, with a multi-dimensional, message-guided fuzzing approach for robotic programs in the Robot Operating System (ROS). By integrating sustainability, AI, and robotics, Song is shaping the future of eco-efficient manufacturing and intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5