Max Robotics — Global Service Robot Directory + US Certification

About

Maciek Chociej is a leading researcher in robotics and reinforcement learning, best known for his groundbreaking work in dexterous manipulation and sim-to-real transfer. His research focuses on enabling robots to perform complex, real-world tasks through simulation-based training, with key contributions in domain randomization and multi-goal learning. Chociej’s most influential work, "Learning dexterous in-hand manipulation" (2019), has garnered over 1,580 citations, demonstrating how reinforcement learning can train a Shadow Dexterous Hand to reorient objects using vision—a milestone in robotic dexterity. He further advanced the field with "Solving Rubik's Cube with a Robot Hand" (632 citations), where he introduced automatic domain randomization (ADR) to bridge the simulation-to-reality gap, enabling a robot hand to solve a Rubik’s Cube entirely from simulated training. Chociej also co-authored "Multi-Goal Reinforcement Learning" (196 citations), which established a suite of challenging robotics benchmarks for pushing, sliding, and in-hand manipulation, widely adopted in the OpenAI Gym ecosystem. His technical work on the OpenAI Remote Rendering Backend (ORRB) has facilitated customizable, high-speed rendering for robotics environments. Chociej’s achievements have set new standards for dexterous manipulation and sim-to-real learning, inspiring a generation of roboticists.