Max Robotics — Global Service Robot Directory + US Certification

About

Taosha Fan is a leading researcher at the intersection of robotics, perception, and differentiable optimization, whose work is shaping how robots achieve human-like dexterity and spatial reasoning. His most influential contribution, the 2024 paper "NeuralFeels with neural fields" (66 citations), pioneers visuotactile perception for in-hand manipulation—enabling robots to estimate an object’s pose and shape by fusing visual and tactile feedback during dexterous tasks. This work directly addresses a core challenge in robotics: inferring spatial awareness from multimodal sensing to reason over contact interactions. Fan is also the creator of Theseus (2022, 47 citations), an open-source library for differentiable nonlinear least squares optimization built on PyTorch. Theseus provides a unified, application-agnostic framework for end-to-end structured learning in robotics and vision, bridging the gap between classical optimization and modern deep learning. His earlier research on online feedback control for input-saturated systems on Lie groups (2016) and efficient higher-order variational integrators (2020) further demonstrates his depth in control theory and simulation. With a growing citation impact and contributions that span perception, learning, and control, Taosha Fan is a rising figure advancing the frontier of autonomous manipulation.