Shinan Huang
Papers
2
Total Citations
53
H-Index
2
About
Shinan Huang is a researcher advancing the frontier of multi-agent coordination, with a primary focus on **Multi-Agent Path Finding (MAPF)** and **Multi-Agent Reinforcement Learning (MARL)**. His most notable contribution is the development of **SCRIMP**, a scalable communication framework that bridges the gap between traditional MAPF algorithms and modern learning-based approaches. By trading off strict performance guarantees for enhanced scalability, SCRIMP enables large teams of agents to learn collaborative, collision-free navigation through reinforcement and imitation learning—a critical step for real-world applications like warehouse robotics and autonomous drone swarms. This work, published in 2023, has already garnered significant attention, accumulating **49 citations** in a short time, reflecting its immediate impact on the field. Huang’s research addresses a key bottleneck in multi-agent systems: how to maintain effective coordination as the number of agents grows. His approach not only improves scalability but also demonstrates how learning-based methods can complement classical planning, opening new avenues for decentralized, communication-efficient pathfinding. For students and researchers, Huang’s work exemplifies the exciting convergence of reinforcement learning and traditional robotics, offering practical tools for tackling complex, real-world multi-agent challenges.
Research Focus
Key Achievements
Top Papers
- 1
- 2