Max Robotics — Global Service Robot Directory + US Certification

About

Weiming Zhi is a robotics researcher whose work spans environmental mapping, motion planning, human-robot interaction, and autonomous navigation. His early contributions tackled fundamental limitations in spatial representation, most notably through continuous occupancy map fusion using Bayesian Hilbert Maps (26 citations), which overcame the discretization constraints of traditional grid-based approaches to give robots richer, more realistic environmental models. Zhi has since pushed into cutting-edge scene reconstruction, developing neural illumination and 3D Gaussian Splatting techniques that enable robots to build photorealistic representations in low-light conditions (13 citations) and removing underwater visual distortions caused by water caustics for seafloor imaging applications. His research into pedestrian trajectory prediction — combining goal-driven and dynamics-based deep learning — directly advances safe autonomous driving and human-aware robot navigation (17 citations), complemented by his SPAN framework for anticipatory crowd navigation. Equally notable is his work democratizing robot instruction through sketch-based learning from demonstration (9 citations) and unifying camera calibration with 3D foundation models. His development of reconfigurable underwater modular robots further demonstrates remarkable breadth. Collectively accumulating over 100 citations, Zhi's research consistently bridges theoretical rigor with practical autonomy challenges across diverse and demanding real-world environments.