Papers
8
Total Citations
70
H-Index
6
About
Yuanheng Zhu is a versatile computational researcher whose work spans reinforcement learning, multi-agent systems, and intelligent control, with a growing reputation for bridging theoretical rigor and real-world robotic applications. His early contributions explored autonomous driving and pedestrian flow optimization, including a notable 2020 study on feedback control in heterogeneous corridors that addressed critical evacuation dynamics, garnering 16 citations. Zhu has since become a prominent voice in cooperative multi-agent reinforcement learning (MARL), co-authoring a widely read 2025 survey on multi-task MARL scenarios with 9 citations, and developing innovative frameworks such as Task-Entity Transformers for handling variable agent configurations across diverse cooperative tasks. His 2024 work stabilizing diffusion models for offline reinforcement learning reflects a keen interest in merging generative AI with robust robotic decision-making. The NeuronsMAE environment he helped create offers a valuable benchmark bridging virtual and physical multi-robot settings. Across morphological policy learning and dynamic target tracking, Zhu consistently pushes the boundaries of adaptive, generalizable robot control. With over 70 cumulative citations and publications spanning top venues, he represents an emerging force in intelligent autonomous systems research.
Research Focus
Key Achievements
Top Papers
- 1Optimal Feedback Control of Pedestrian Flow in Heterogeneous Corridors16 citations · 2020
- 2
- 3
- 4
- 5
- 6
- 7
- 8