Max Robotics — Global Service Robot Directory + US Certification

About

Martijn J.A. Zeestraten is a leading researcher in robot learning and control, specializing in imitation learning, movement primitives, and human-robot interaction. His work addresses fundamental challenges in enabling robots to learn complex tasks from human demonstration, with a particular focus on representing and controlling orientation and motion in non-Euclidean spaces. Zeestraten’s most influential contribution is his 2017 paper on imitation learning on Riemannian manifolds (109 citations), which introduced a principled way to encode end-effector orientation using Riemannian geometry—overcoming the limitations of standard Euclidean approaches. He further advanced the field with his work on variable duration movement encoding and minimal intervention control (18 citations), enabling robots to adapt motion timing while maintaining task performance. His 2018 paper on flexible automation driven by demonstration (22 citations) highlights the practical implications of his research for simplifying robotics in industrial settings. Zeestraten has also made notable contributions to safe human-robot collaboration, proposing frameworks for online motion synthesis with formal safety guarantees (10 citations). His research on learning task-space synergies using Riemannian geometry (6 citations) continues to influence how robots achieve coordinated, task-specific behaviors. With a strong focus on both theoretical rigor and real-world applicability, Zeestraten’s work is essential reading for researchers in robot learning, control, and human-robot collaboration.