Tianhu Peng

University of Leeds, University College London

Papers

Total Citations

H-Index

About

Tianhu Peng is a rising researcher at the intersection of robotics and artificial intelligence, with a primary focus on bipedal locomotion and reinforcement learning. His work addresses the fundamental challenge of enabling robots to walk with human-like stability and adaptability. Peng’s most notable contribution, “Learning Bipedal Walking on a Quadruped Robot via Adversarial Motion Priors” (2024), demonstrates a groundbreaking approach: repurposing a quadruped platform to achieve bipedal gait through adversarial training, effectively decoupling morphology from locomotion policy. This work has already garnered 8 citations, signaling its early impact. Complementing this, his survey “Deep reinforcement learning for robotic bipedal locomotion: a brief survey” (2025) synthesizes the fragmented DRL landscape, identifying critical gaps toward a unified framework for real-world deployment. With 7 citations, this review serves as a vital resource for researchers navigating the field. Peng’s contributions are particularly timely as the robotics community seeks robust, transferable locomotion controllers. His innovative use of adversarial motion priors and his systematic mapping of DRL challenges position him as a key voice in advancing bipedal robotics from simulation to practical application.