Max Robotics — Global Service Robot Directory + US Certification

About

Qimeng Tan’s research advances the frontier of autonomous mobile robotics, with a focus on robust perception, localization, and dynamic manipulation. Their work tackles fundamental challenges in enabling robots to navigate and interact reliably in complex, changing environments. Tan’s key contributions span multi-sensor fusion for navigation, where a method using ultrasonic sensors to build global maps for path planning was developed (14 citations). In visual localization, Tan introduced the 2-Entity RANSAC algorithm, a robust approach that maintains accuracy despite environmental changes (11 citations), and later achieved globally optimal consensus maximization for visual-inertial localization using point and line maps (7 citations). Addressing appearance changes over time, Tan proposed explicit feature disentanglement for visual place recognition (4 citations). Notably, Tan’s work on neural motion prediction enables robots to catch uneven, in-flight objects by predicting complex trajectories within milliseconds (12 citations). This body of work, spanning from foundational fuzzy matching for localization (2002) to cutting-edge deep learning methods, demonstrates a sustained impact on practical robot autonomy and perception.