Papers
2
Total Citations
17
H-Index
2
About
Xiangyu Ning is a roboticist advancing the frontiers of autonomous navigation and skill acquisition through imitation and reinforcement learning. His research centers on enabling mobile robots to learn complex behaviors safely and efficiently from human demonstrations. In a seminal 2019 work, Ning pioneered the use of Dynamic Movement Primitives (DMPs) within a Nonlinear Model Predictive Control framework, allowing robots to fluidly imitate human trajectories—a method that reduces search complexity and makes human-robot interaction more intuitive. This paper has garnered 11 citations for its practical approach to learning by demonstration. More recently, in 2022, Ning tackled the critical challenge of safe autonomy with a novel Constrained Hierarchical Reinforcement Learning algorithm, achieving 6 citations by integrating safety constraints directly into the learning process for high-dimensional control. His work bridges the gap between human-guided learning and robust, real-world deployment, offering a pathway toward robots that can navigate unpredictable environments without compromising safety. Ning’s contributions are particularly notable for their focus on hierarchical structures that decompose complex tasks, making his methods both scalable and applicable to real robotic systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2