Max Robotics — Global Service Robot Directory + US Certification

About

Gabriele Tiboni is a researcher at the forefront of robotic manipulation, specializing in bridging the sim-to-real gap through advanced domain randomization techniques. His work focuses on enabling reinforcement learning policies to transfer seamlessly from simulation to physical robots, particularly in the challenging domain of soft robotics. Tiboni’s major contribution is the development of DROPO, a novel method for offline domain randomization that optimizes randomization distributions without requiring online interaction, making sim-to-real transfer more robust and accessible. This work has garnered 33 citations since 2023, reflecting its significant impact on the field. Additionally, his research on domain randomization for closed-loop control of soft robots addresses the fundamental challenge of modeling systems with infinite degrees of freedom, demonstrating how robust policies can be learned despite highly approximated dynamics. By systematically comparing online and offline adaptation strategies, Tiboni provides crucial benchmarks that guide practitioners in selecting appropriate methods for real-world deployment. His work is instrumental in making soft robots—valued for their safety and adaptability—practical for real-world applications through affordable, effective control.