Max Robotics — Global Service Robot Directory + US Certification

About

Mingyang Geng is a robotics and artificial intelligence researcher whose work sits at the intersection of multi-robot systems, deep learning, and autonomous navigation. His most influential contributions focus on enabling decentralized multi-robot cooperation through learned communication strategies, most notably his development of attention-based communication neural networks that allow robots to coordinate exploration of unknown environments without relying on a centralized controller. This line of research, spanning foundational work in 2018 and a refined follow-up in 2019, has garnered over 40 citations combined, establishing him as a meaningful voice in the multi-agent reinforcement learning community. Beyond exploration, Geng has advanced the field of autonomous trail following, demonstrating how sensor fusion and deep learning can be combined to extend single-robot navigation capabilities to collaborative multi-robot teams operating in challenging outdoor environments. His additional work on model partitioning addresses a critical practical barrier in robotics: deploying computationally expensive neural networks on resource-constrained hardware through crowdsourcing-style distributed inference. Taken together, Geng's research reflects a consistent drive to bridge the gap between theoretical deep learning advances and real-world, resource-limited robotic deployment, making autonomous multi-robot systems more intelligent, adaptive, and practically viable.