Max Robotics — Global Service Robot Directory + US Certification

About

Ritvik Singh is at the forefront of bridging the sim-to-real gap in robotic manipulation, with a focus on dexterous, multi-fingered systems and scalable simulation environments. His landmark work, *Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments* (226 citations), established a modular, high-fidelity platform powered by NVIDIA Isaac Sim, enabling researchers to efficiently create photo-realistic scenes for training and evaluating robot learning algorithms. This framework has become a cornerstone for the community, accelerating the development of robust policies. Singh’s research is perhaps best exemplified by *DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality* (88 citations), where he demonstrated that complex, agile manipulation skills—like spinning a pen or reorienting objects—can be learned entirely in simulation and successfully transferred to a real-world robotic hand. This work showcased a breakthrough in overcoming the notorious simulation-to-reality gap. More recently, with *Synthetica: Large Scale Synthetic Data Generation for Robot Perception* and *DextrAH-RGB: Visuomotor Policies to Grasp Anything with Dexterous Hands*, he is pushing towards generalist, vision-driven grasping systems that operate without relying on depth maps or object poses. Singh’s contributions are defining the next generation of scalable, data-driven robotics.