Dynamo is a cognitive robotic system built to adapt to unstructured environments and learn through interaction. By integrating foundation models with rigorous control theory, it achieves versatile, robust behavior in real-world settings.

A hierarchical pick-and-place pipeline built with LeRobot: a planner grounds which vial goes where and emits a source-to-destination command, then the flow policy executes the spatial move from three RGB views, inferring on the Jetson beside the arm. Evaluation across autonomous rollouts localizes the limit at grasping rather than at language or targeting.

The Ridgeback picks one designated person out of a multi-person scene, then fuses camera detection with point-cloud perception to follow them through clutter. Two iterations compare 2D and full 3D sensing for environment understanding.