Max Robotics — Global Service Robot Directory + US Certification

About

Josiah Wong is a leading researcher in embodied AI and robot manipulation, with a focus on bridging the gap between simulation and real-world robotic learning. His most impactful contribution is the development of **iGibson 1.0**, a high-fidelity simulation environment featuring 15 large-scale, fully interactive home scenes with 108 rooms. This platform, which has garnered over 177 combined citations, has become a cornerstone for training and evaluating robots on complex, interactive tasks like navigation and object manipulation. Wong’s work on **learning from offline human demonstrations** (70 citations) has critically advanced imitation learning and offline reinforcement learning for manipulation, identifying key factors that make human data effective for training robots. He has also pioneered methods for **multi-arm manipulation** through collaborative teleoperation and introduced **OSCAR**, a data-driven approach to operational space control that enhances robot adaptability and robustness. More recently, Wong contributed to **BEHAVIOR-1K**, a massive benchmark of 1,000 everyday activities designed to push embodied AI toward human-centered applications. His research consistently emphasizes scalable, realistic simulation and data-efficient learning, making him a pivotal figure in the quest for capable, general-purpose household robots.