Abstract: “Current AI systems for perception and action incorporate a number of techniques: optimal observer models, Bayesian filtering, probabilistic mapping, trajectory planning, dynamic navigation, and feedback control. I will describe and demonstrate some of these methods for autonomous driving and legged and flying robots, and contrast these models with neural representations and computation in biology. I will also highlight some new research on machine learning for these systems, and discuss the role of geometrical structures and noise in synthetic and biological approaches to classification and decision making.”