Max Robotics — Global Service Robot Directory + US Certification

About

Xinjing Cheng is a robotics and computer vision researcher whose work centers on autonomous robot navigation and omnidirectional perception — two areas increasingly critical to the deployment of intelligent mobile systems in real-world environments. His most influential contributions tackle one of robotics' most persistent challenges: enabling robots to move safely and purposefully through dense, unpredictable human crowds. His 2019 paper, "Getting Robots Unfrozen and Unlost in Dense Pedestrian Crowds," has accumulated 68 citations and directly addresses the notorious "freezing robot problem," proposing strategies that allow robots to make meaningful progress in environments such as shopping malls and transit terminals. Complementing this, his 2018 work on CrowdMove (47 citations) introduced a compelling multi-robot, multi-scenario training framework that generalizes mapless navigation across diverse crowded settings using robust policy gradient methods. Beyond navigation, Cheng has made noteworthy strides in omnidirectional depth perception, developing CNN-based architectures that extend depth information from standard sensors to full 360° fields of view, addressing a key hardware limitation for autonomous robots. Collectively amassing over 140 citations, his research offers both theoretical insight and practical solutions for next-generation autonomous systems operating alongside humans.