Max Robotics — Global Service Robot Directory + US Certification

About

Armin Biess is a researcher working at the intersection of robotics, machine learning, and autonomous systems, with a particular focus on robot learning, motion planning, and human-robot interaction. His work spans several compelling areas, including reinforcement learning for collaborative robotics, pose estimation for industrial assembly, and data-driven approaches to modeling human behavior. Among his most notable contributions, Biess has investigated the application of Guided Policy Search — a data-efficient model-based reinforcement learning method — to the challenging problem of human-robot object handovers, a fundamental capability for collaborative robots operating alongside humans. He has also advanced flexible robotic assembly by leveraging simulated depth images and 3D CAD models to train pose estimation systems, reducing the need for rigid initial conditions in industrial settings. More recently, his work on Global Tensor Motion Planning addresses the growing demand for scalable, batch motion planning to support downstream learning applications such as imitation learning. His research on example-guided deep reinforcement learning for modeling stochastic human driving policies further demonstrates his broad interest in learning-based autonomous behavior. With citations accumulating across robotics and AI venues, Biess represents a researcher steadily shaping the future of intelligent, adaptable robotic systems.