Jinming Xu

Papers

Total Citations

H-Index

About

Jinming Xu is a rising researcher at the forefront of multi-agent systems and reinforcement learning (RL), with a focus on bridging the critical sim-to-real gap in robotic exploration. His most-cited work, "MAexp: A Generic Platform for RL-based Multi-Agent Exploration" (2024, 4 citations), introduces a novel framework that tackles the inefficiencies of scene quantization and action discretization—key barriers to deploying RL in real-world multi-agent scenarios. By providing a unified, generic platform, Xu enables diverse MARL algorithms to be tested and compared more effectively, directly addressing the lack of diversity and sampling inefficiency in existing simulation environments. This contribution is foundational for advancing autonomous exploration in robotics, from search-and-rescue to planetary mapping. Though early in his career, Xu’s work has already garnered attention for its practical approach to a notoriously difficult problem, positioning him as a promising voice in multi-agent decision-making and sim-to-real transfer. His research holds significant potential for students and engineers seeking to deploy robust, scalable RL solutions in complex, unstructured environments.