Xintong Yang

Cardiff University

Papers

7

Total Citations

157

H-Index

5

About

Xintong Yang is a roboticist pushing the boundaries of autonomous manipulation and navigation through hierarchical reinforcement learning (HRL). Her research centers on enabling robots to perform complex, multistep tasks—such as block stacking, assembly, and mapless navigation—by combining low-level motion control with high-level symbolic planning. Her most cited work, “Hierarchical Reinforcement Learning With Universal Policies for Multistep Robotic Manipulation” (2021, 88 citations), introduces a framework that allows robots to efficiently learn long-horizon tasks with sparse rewards, a critical challenge in robotics. Yang also contributed the open-source multi-goal RL environment for PyBullet (2021), a widely used resource for benchmarking robotic manipulation. Her 2023 survey on deep robotic affordance learning from an RL perspective (20 citations) synthesizes key advances in how robots perceive and act upon objects. More recently, she has explored differentiable physics for manipulating deformable materials like dough and clay (2025), and human-robot collaboration through automated task knowledge generation (2022). With over 150 total citations, Yang’s work is shaping the next generation of intelligent, adaptable robots capable of operating in unstructured environments.

Research Focus

Key Achievements

5
H-Index
7
Papers
157
Total Citations
22
Avg Citations/Paper
🏆 Most Cited Paper
Hierarchical Reinforcement Learning With Universal Policies for Multistep Robotic Manipulation
88 citations · 2021
📈 Most Prolific Year: 2021 (2 Papers)
🤝 Key Collaborators: 9
🏛 Institutions: Cardiff University

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago