Max Robotics — Global Service Robot Directory + US Certification

About

Brian D. Ziebart is a leading researcher in robotics and artificial intelligence, specializing in human-robot interaction, inverse optimal control, and imitation learning. His most influential work, "Planning-based prediction for pedestrians" (2009, 469 citations), pioneered the use of maximum entropy inverse optimal control to model goal-directed human trajectories, enabling robots to navigate crowded environments without hindering people. This foundational contribution has shaped how robots predict and adapt to human behavior. Ziebart’s research extends to intent prediction and trajectory forecasting, where he developed predictive inverse linear-quadratic regulation (2015) to infer human intentions from noisy sensor data, and graph-based inverse optimal control (2015) for robot manipulation. He has also advanced robotic teleoperation through deep correspondence learning (2019) and tackled covariate shift in imitation learning (2021), identifying three critical regimes for policy stability. His work on risk-averse policy optimization (2022) and learning from video demonstrations (2023) further demonstrates his commitment to safe, efficient robot learning. With over 600 total citations, Ziebart’s contributions are essential for students and researchers seeking to build robots that seamlessly collaborate with humans in dynamic, real-world environments.