Max Robotics — Global Service Robot Directory + US Certification

About

Johannes Tenhumberg is a roboticist advancing the frontier of autonomous manipulation through deep learning and self-contained calibration. His research centers on three interconnected challenges: enabling high-DOF robots to plan collision-free motions in complex environments, achieving precise kinematic calibration without external sensors, and accelerating inverse kinematics for real-time control. His most impactful work, “Speeding Up Optimization-based Motion Planning through Deep Learning” (14 citations), pioneers the use of neural networks to encode prior motion planning experience, dramatically reducing computation time for robots navigating cluttered spaces. Tenhumberg has also developed innovative self-calibration methods, including a head-mounted RGB camera system for elastic humanoid upper bodies (7 citations) and a pairwise contact technique for multi-fingered hands (5 citations)—both eliminating the need for expensive external tracking equipment. His 2021 paper on calibrating joint and transversal elasticities in humanoid upper bodies (7 citations) was the first to model these flexibilities for improved absolute accuracy in manipulation tasks. By combining learning-based speedups with practical, sensor-free calibration, Tenhumberg’s work brings humanoid robots closer to robust, autonomous operation in real-world settings.