Papers
10
Total Citations
267
H-Index
6
About
Yandong Ji is a robotics researcher whose work spans legged locomotion, whole-body control, and human-robot interaction, with growing influence across both academic and applied robotics communities. His research has made significant strides in enabling quadrupedal and humanoid robots to perform complex real-world tasks through reinforcement learning. Most notably, his work on expressive whole-body control for humanoid robots (72 citations) demonstrates how large-scale human motion capture data can drive lifelike robot movement, while DribbleBot (43 citations) and his hierarchical reinforcement learning framework for soccer shooting (58 citations) showcase legged robots performing dynamic, dexterous manipulation in unstructured environments. Beyond locomotion, Ji has contributed meaningfully to human augmentation, with his study on ankle exoskeleton-assisted walking (62 citations) informing the design of more effective wearable rehabilitation devices. His more recent contributions—including NaVILA, a vision-language-action model for legged robot navigation, and RoboDuet, a cooperative loco-manipulation framework—reflect a forward-looking trajectory toward language-guided, multi-modal robotic systems. Collectively accumulating over 260 citations, Ji's body of work positions him as an emerging force in intelligent legged robotics.
Research Focus
Key Achievements
Top Papers
- 1Expressive Whole-Body Control for Humanoid Robots72 citations · 2024
- 2
- 3
- 4DribbleBot: Dynamic Legged Manipulation in the Wild43 citations · 2023
- 5NaVILA: Legged Robot Vision-Language-Action Model for Navigation13 citations · 2025
- 6
- 7
- 8Expressive Whole-Body Control for Humanoid Robots2 citations · 2024
- 9Visual Whole-Body Control for Legged Loco-Manipulation2 citations · 2024
- 10NaVILA: Legged Robot Vision-Language-Action Model for Navigation2 citations · 2024