Xionglei Zhao
Papers
2
Total Citations
15
H-Index
2
About
Xionglei Zhao is a researcher advancing the field of robotic manipulation through the integration of computer vision and deep learning. His work centers on enabling robots to perceive, recognize, and grasp objects in complex, unstructured environments—a critical challenge for autonomous systems. In his highly cited 2018 paper, "A Vision-Based Robotic Grasping Approach under the Disturbance of Obstacles" (10 citations), Zhao introduced a method that combines deep learning-based object detection with Euclidean cluster segmentation to allow robots to successfully grasp targets even when obstacles are present. Building on this, his subsequent work, "Object Recognition, Localization and Grasp Detection Using a Unified Deep Convolutional Neural Network with Multi-task Loss" (5 citations), proposed a unified CNN architecture that simultaneously handles object recognition, localization, and grasp detection from RGB-D data. By representing grasps as two-point coordinates and employing a multi-task loss function, this approach streamlines the perception-to-action pipeline. Zhao’s contributions are foundational for developing more adaptive and reliable robotic systems, with direct applications in manufacturing, logistics, and service robotics. His research continues to push the boundaries of how machines interact with the physical world.
Research Focus
Key Achievements
Top Papers
- 1A Vision-Based Robotic Grasping Approach under the Disturbance of Obstacles10 citations · 2018
- 2