Perception

Perception transforms raw sensing into map, localization, and task-relevant scene understanding. The chapter emphasizes probabilistic mapping and vision-in-the-loop feedback.

Core Questions

Which state representations remain observable with sparse landmarks?
How should map confidence be accumulated and decayed?
How do perception delays impact closed-loop control?

Algorithms

Prerequisites

Probabilistic robotics fundamentals
Camera projection and feature tracking
Grid-map and landmark-map representations