Byoung-Tak Zhang

Papers

Total Citations

H-Index

About

Byoung-Tak Zhang is a leading researcher in robotics and artificial intelligence, with a focus on evolutionary computation and perception systems for autonomous manipulation. His pioneering work includes developing methods to evolve behavior-based robot controllers using genetic programming, enabling robots to learn complex tasks hierarchically. This foundational research, cited over 18 times, addresses the challenge of incorporating sensory inputs into high-level knowledge representation for robotic behavior. Zhang has made significant contributions to robotic perception in cluttered environments, notably through his work on Multi-Object RANSAC, an efficient plane clustering method for RGB-D cameras that enhances robot grasping in complex scenes. His recent research on DA-Fusion introduces a deformable attention-based RGB-D fusion transformer for unseen object instance segmentation, achieving precise segmentation critical for logistics automation tasks like bin-picking and shelf-picking. This work tackles challenges such as occlusions and varying object shapes, demonstrating robust perception in real-world applications. With a career spanning over two decades, Zhang’s research has been cited in numerous studies, reflecting its impact on evolutionary robotics and computer vision. His innovative approaches continue to advance the field, making him a notable figure in developing intelligent robotic systems for dynamic environments.