Max Robotics — Global Service Robot Directory + US Certification

About

Oleg Arenz is a robotics and machine learning researcher whose work spans robotic teleoperation, imitation learning, and physics-informed control. His most recognized contribution, "A Haptic Shared-Control Architecture for Guided Multi-Target Robotic Grasping" (2019, 71 citations), addresses a long-standing challenge in nuclear remote handling by moving beyond outdated master-slave manipulator paradigms toward intelligent, haptic-guided robotic grasping systems. This work exemplifies his broader interest in making robotic manipulation more intuitive and capable in demanding real-world environments. Arenz has also advanced the field of assisted teleoperation through trajectory learning methods that remain robust even when training demonstrations are suboptimal — a practical and often overlooked problem in human-robot interaction. His theoretical contributions extend into imitation learning, where he explored non-adversarial alternatives to popular GAN-based approaches like GAIL, clarifying their mathematical connections and offering more stable training frameworks. More recently, he has pursued physics-consistent deep learning for model predictive control, developing context-aware Lagrangian networks capable of handling complex, object-rich environments. Across these areas, Arenz consistently bridges rigorous theoretical foundations with practical robotic applications, making his work valuable to both researchers and engineers in the field.