Xiangli Nie
Papers
2
Total Citations
10
H-Index
2
About
Xiangli Nie is a roboticist whose research lies at the intersection of continual learning, autonomous perception, and bio-inspired control. Her most influential work addresses the critical challenge of lifelong object recognition for robots operating in dynamic, real-world environments. In her 2023 paper "Online Active Continual Learning for Robotic Lifelong Object Recognition" (7 citations), Nie proposes a framework that enables robotic systems to continuously and interactively learn from streaming data without catastrophic forgetting—a fundamental hurdle for truly autonomous agents. This work is pivotal for deploying robots that must adapt to novel objects and changing contexts over extended periods. Additionally, Nie tackles the complexity of musculoskeletal robots in her 2021 study "Redundancy Reduction of Musculoskeletal Model for Robots with Group Sparse Neural Network" (3 citations), where she introduces a heuristic approach using group sparse neural networks to simplify the highly redundant muscle systems that make such robots difficult to control and manufacture. Her contributions bridge the gap between theoretical machine learning and practical robotics, offering scalable solutions for lifelong adaptation and efficient bio-inspired design.
Research Focus
Key Achievements
Top Papers
- 1Online Active Continual Learning for Robotic Lifelong Object Recognition7 citations · 2023
- 2