Weihang Liang
Papers
4
Total Citations
198
H-Index
2
About
Weihang Liang is a robotics researcher specializing in autonomous navigation, human-robot interaction, and deep reinforcement learning. Their work centers on one of mobile robotics' most pressing challenges: enabling robots to navigate safely and efficiently through dynamic, crowded human environments. Liang's most influential contribution, "Decentralized Structural-RNN for Robot Crowd Navigation with Deep Reinforcement Learning" (118 citations), addresses critical limitations in prior navigation systems by tackling partial observability and unknown agent dynamics — real-world conditions that earlier methods struggled to handle. This work introduced a decentralized structural approach using recurrent neural networks to model complex multi-agent interactions more robustly. Building on this foundation, Liang's subsequent research on intention-aware navigation (76 citations) advanced the field further by incorporating attention-based interaction graphs that capture diverse agent relationships and anticipate human intentions — significantly improving robot decision-making in dense, interactive crowds. Together, these contributions have accumulated nearly 200 citations, reflecting meaningful influence within the robotics and AI communities. Liang's trajectory demonstrates a consistent commitment to bridging the gap between theoretical reinforcement learning methods and the messy, unpredictable realities of human environments — work with direct implications for service robots, autonomous vehicles, and assistive technologies.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4