Guanyu Gao

Nanjing University of Science and Technology, Nanyang Technological University

Papers

Total Citations

H-Index

About

Guanyu Gao is a researcher advancing the frontiers of social robot navigation and model-based reinforcement learning (MBRL). His work addresses critical challenges in enabling robots to operate safely and naturally in crowded human environments. Gao’s key contributions include the development of **SocialGAIL** (2024, 5 citations), a generative adversarial imitation learning framework that produces faithful crowd simulations for training social robots. This approach overcomes limitations of traditional reinforcement learning by generating realistic, diverse pedestrian behaviors, significantly improving robot navigation policies in dense spaces. In parallel, Gao introduced **Baconian** (2019, 2 citations), a unified open-source framework for MBRL that standardizes and simplifies the implementation of dynamics models and planning algorithms. This tool has accelerated research in robotics and autonomous driving by reducing engineering overhead. Gao’s work bridges the gap between simulation fidelity and real-world deployment, offering practical solutions for human-robot interaction. His contributions are particularly notable for their emphasis on reproducibility and accessibility, making advanced RL techniques more approachable for the broader research community.