Active Observation and Visual Servoing
IEEE CASE 2026 · Accepted · First-author paper
During a grasp, the robot arm or gripper can obscure the target just when reliable visual feedback is most important. This work uses one arm to position an RGB-D camera while another performs the grasp.
The method combines a two-stage viewpoint strategy, target–gripper relative image error, and occlusion-triggered viewpoint exploration to maintain or recover useful visual feedback.
Paper
Active Observation for Robust Visual Servoing in Robotic Manipulation
Zhilun Hu, Yuqi Cheng, Yuxiang Tan, Wen-pan Li, Weiming Shen.
IEEE International Conference on Automation Science and Engineering (CASE), 2026.