Max Robotics — Global Service Robot Directory + US Certification

About

Matthias Plappert is a leading researcher in reinforcement learning (RL) and robotic manipulation, renowned for bridging the gap between simulation and real-world dexterity. His seminal work, *"Learning dexterous in-hand manipulation"* (1,588 citations), pioneered the use of domain randomization to train vision-based policies that enable a physical Shadow Dexterous Hand to reorient objects—a feat previously considered intractable. He further advanced this paradigm in *"Solving Rubik's Cube with a Robot Hand"* (632 citations), introducing **automatic domain randomization (ADR)** to solve a manipulation problem of unprecedented complexity entirely in simulation before zero-shot transfer to a real robot. Plappert also contributed foundational infrastructure to the field, co-authoring the *"Multi-Goal Reinforcement Learning"* report (196 citations), which introduced the widely-adopted Fetch and Shadow Hand environments within OpenAI Gym. His work on asymmetric self-play for automatic goal discovery and tactile sensing for sample efficiency continues to push the boundaries of what RL can achieve in robotics. With over 2,500 total citations, Plappert’s research has become a cornerstone for anyone working on sim-to-real transfer, dexterous manipulation, and goal-conditioned RL.