Linrui Tian

Papers

Total Citations

H-Index

About

Linrui Tian is a researcher advancing the frontiers of robotic manipulation and autonomous control in challenging, real-world environments. Her work centers on two key areas: intelligent control for mobile robotics and robust robotic grasping under perceptual uncertainty. In her highly cited 2020 paper, "Fast neural network control of a pseudo-driven wheel on deformable terrain," Tian developed a neural network-based control strategy to enable stable and efficient locomotion on unpredictable, yielding surfaces—a critical problem for field robotics. This work has garnered 13 citations, reflecting its importance for off-road and planetary rover applications. Tian’s most notable contribution, however, is the Grasp-Vote Network (GVN), introduced in her 2021 paper "Vote for Grasp Poses from Noisy Point Sets by Learning From Human." Addressing the persistent challenge of robotic grasping from noisy, incomplete point clouds—such as those from reflective or cluttered scenes—GVN learns from human demonstrations to directly vote for feasible grasp poses, achieving robust performance where conventional methods fail. This innovative approach to learning from human input for perception-driven manipulation marks a significant step toward more reliable and adaptable robotic hands.