An energy-based perspective on learning observation models

Robots perceive the rich world around them through the lens of their sensors. Each sensor observation is a tiny window into the world that provides only a partial, simplified view of reality. To complete their tasks, robots combine multiple readings from sensors into an internal task-specific representation of the world that we call state. This internal picture of the world enables robots to evaluate the consequences of possible actions and eventually achieve their goals

articles deep dive energy learning perspective