Papers
11
Total Citations
154
H-Index
4
About
Mengdi Xu is a researcher at the forefront of robotics and machine learning, with key contributions spanning biomachine hybrid robots, flexible sensors, and robust reinforcement learning. Their work on energy supply for biomachine hybrid robots (68 citations) addresses critical challenges in deploying robots for wilderness rescue and ecological monitoring. Xu also pioneered extrusion printing of carbon nanotube-coated elastomer fibers for flexible pressure sensors (41 citations), advancing wearable and soft robotics applications. In reinforcement learning, Xu introduced a novel framework treating robust RL as a Stackelberg game via adaptively-regularized adversarial training (22 citations), significantly improving agent performance under model errors and adversarial attacks. Notable recent work includes leveraging large language models for creative robot tool use and closing the sim-to-real gap through differentiable causal discovery. Xu’s research on safe falling for quadrupedal robots and interactive camera calibration further demonstrates their breadth. With over 150 total citations and publications in top venues, Xu’s work is shaping the future of resilient, intelligent robotic systems.
Research Focus
Key Achievements
Top Papers
- 1A Review of Energy Supply for Biomachine Hybrid Robots68 citations · 2023
- 2
- 3
- 4
- 5
- 6Guardians as You Fall: Active Mode Transition for Safe Falling3 citations · 2024
- 7Creative Robot Tool Use with Large Language Models3 citations · 2023
- 8
- 9
- 10