Qingni Hu

Papers

Total Citations

112

H-Index

About

Qingni Hu is a pioneering researcher in robotics and neural computation, whose work has fundamentally advanced the kinematic control of redundant robot manipulators. Her key research areas include recurrent neural networks, real-time optimization, and redundancy resolution in robotic systems. Hu’s major contribution lies in developing innovative neural network architectures that solve complex inverse kinematics problems with remarkable efficiency. Her seminal 1999 paper, “A Lagrangian network for kinematic control of redundant robot manipulators,” which has garnered 104 citations, introduced a groundbreaking recurrent neural network that determines optimal redundancy resolution through real-time quadratic optimization. This work demonstrated how neural networks could replace traditional computational methods for instantaneous control of robotic arms. Building on this foundation, her 2002 paper proposed a two-layer recurrent neural network architecture with bidirectionally connected neuron arrays, enabling more sophisticated kinematic control by processing end-effector velocity signals directly. Hu’s research bridges the gap between neural computation theory and practical robotics applications, offering elegant solutions to the challenge of coordinating multiple degrees of freedom in robotic manipulators. Her work continues to influence modern approaches to real-time robot control and neural-based optimization in automation systems.