Max Robotics — Global Service Robot Directory + US Certification

About

Xiaobao Tong is a researcher advancing the frontier of robotic perception and manipulation, with a core focus on multifinger grasping, tactile object recognition, and cross-modal sensory fusion. Tong’s major contributions lie in developing adaptive, kernel-based learning methods that preserve the rich, multidimensional structure of tactile data—overcoming the limitations of traditional approaches that flatten force signals across fingers. For instance, their work on Adaptive Multikernel Dictionary Learning (22 citations) explicitly models inter-finger force coupling, enabling more accurate grasping state recognition. Tong has also pioneered few-shot and cross-modal learning for robotics, as seen in their multispectral few-shot coupled learning framework (19 citations), which allows robots to recognize unknown objects with minimal training data. Further, their variational Bayesian Gaussian mixture model for cross-modal generation (5 citations) addresses the challenge of poor-quality synthetic data in multimodal fusion. While early in their career, Tong’s work is already shaping how robots perceive and interact with the physical world—bridging tactile, visual, and spectral data to achieve more human-like dexterity and object understanding.