Guangzhe Zhao

Yanbian University, Beijing University of Civil Engineering and Architecture

Papers

Total Citations

H-Index

About

Guangzhe Zhao is a robotics researcher whose work bridges machine learning, human-robot interaction, and computer vision. His primary research areas include robot learning from demonstration, reinforcement learning, and lightweight deep learning architectures for humanoid robotics. Zhao’s major contribution lies in developing adaptive robot control methods that combine locally weighted regression (LWR) with Q-learning algorithms, enabling robots to learn complex tasks through demonstration and generate novel actions—demonstrated effectively on a 6-DOF hitting-ball system. This work, published in 2018, has garnered 4 citations and represents a foundational approach to skill transfer in robotics. Additionally, Zhao has advanced human-robot interaction through lightweight convolutional neural networks (CNNs) for real-time facial expression recognition on humanoid robots, addressing the critical challenge of deploying deep learning models on resource-constrained platforms. This 2020 work, with 2 citations, tackles the trade-off between accuracy and computational efficiency in embedded robotic systems. Zhao’s research is particularly notable for its practical focus on enabling robots to perceive human emotional states—such as pain and fatigue—through efficient visual processing, a capability essential for safe and intuitive human-robot collaboration. His work continues to influence the development of adaptive, perceptive robotic systems.