Building Real-Time, Adaptive Robots Using Online Search

Title: Building Real-Time, Adaptive Robots Using Online Search

Abstract: Robots operating in unpredictable environments often struggle when offline-trained policies—particularly those developed in simulation—face domain shifts. Although techniques like domain-randomized reinforcement learning and imitation learning have shown impressive results in hardware, they typically require extensive retraining and are unable to adapt to changing conditions. In this talk, I introduce an alternative approach that leverages online search to adapt pre-trained policies at runtime, allowing robots to meet new constraints and optimize new objectives without expensive retraining. I will first present our work on stochastic safety, which employs a control barrier function calibrated with a hardware-trained generative model to enable real-time collision avoidance for quadrotors and humanoid robots. I will also discuss our recent work online search for dexterous manipulation, where forward search via the cross-entropy method and vision-based state estimation achieve robust in-hand cube reorientation in hardware, without any policy pretraining. These findings demonstrate that runtime computation and online search can enable robust adaptation and multi-task generalization, opening promising new research directions in adaptive robotics.

Bio: Preston Culbertson is an incoming Assistant Professor of Computer Science at Cornell University, starting in Fall 2025. His research focuses on closing the gap between robotic capabilities and real-world complexity, developing robots that move, manipulate, and adapt with true robustness. Drawing on ideas from machine learning, computer vision, and control theory, his work explores adaptable strategies for dexterous manipulation and dynamic locomotion. Preston is currently a Research Scientist at the Robotics and AI Institute, and was previously a postdoctoral scholar at Caltech. He earned his Ph.D. in Mechanical Engineering from Stanford in 2022.