Wenli Xiao

Papers

Total Citations

H-Index

About

Wenli Xiao is pioneering the frontier of human-robot interaction and adaptive autonomy, with a focus on humanoid teleoperation, safe policy adaptation, and agile mobility. Their most impactful work, "Learning Human-to-Humanoid Real-Time Whole-Body Teleoperation" (H2O), has garnered 68 citations and introduces a reinforcement learning framework that enables real-time, whole-body control of a full-sized humanoid robot using only an RGB camera. This breakthrough includes a scalable method for creating large-scale retargeted motion datasets, bridging the gap between human movement and robotic embodiment. In "Safe Deep Policy Adaptation" (8 citations), Xiao addresses the critical need for autonomous systems to rapidly adapt in dynamic environments while maintaining safety and stability, combining insights from classic adaptive control with modern policy learning. Their recent work, "AnyCar to Anywhere" (6 citations), advances universal dynamics models for agile and adaptive mobility across diverse robotic platforms, pushing the limits of performance in navigation and locomotion. Xiao’s research stands at the intersection of reinforcement learning, safe control, and embodied AI, offering transformative tools for real-world robotic deployment. Their contributions are shaping the next generation of autonomous systems that are both highly capable and reliably safe.