Max Robotics — Global Service Robot Directory + US Certification

About

Qiangfu Zhao is a researcher whose work lies at the intersection of robotics, computer vision, and efficient deep learning hardware. His key research areas include obstacle avoidance for autonomous robots, neural network-based control systems, and hardware acceleration for convolutional neural networks (CNNs). A major contribution is his investigation of depth image-based obstacle avoidance for indoor patrol robots, addressing the long-standing challenge of lighting dependence in vision-based systems—a study that has garnered 9 citations. Zhao has also advanced the field of efficient AI inference by proposing a random-forest-based approximation layer unit (RFA-LU) for binary and ternary CNN accelerators, aiming to enable low-latency, low-power robot control. This work, with 4 citations, demonstrates his focus on practical, area-efficient hardware solutions. Earlier in his career, Zhao explored the generation of smart robot controllers through co-evolution and the extraction of decision trees from evolved neural network controllers, contributing to the interpretability and training of autonomous systems. His research consistently bridges the gap between theoretical machine learning and real-world robotic applications, making him a notable figure in the development of intelligent, resource-constrained robotic systems.