A Beginner's Glossary of Perception Sensor Terms

The vocabulary that shows up in almost every discussion of how robots sense the world.

The basics

Field of view (FoV): the angular extent of the world a sensor can see at once. Point cloud: a set of 3D coordinates representing sensed surfaces, the typical output of LiDAR and depth cameras. Odometry: estimating a robot's change in position over time from sensor data, without relying on an external reference like GPS.

A bit further in

Extrinsic calibration: determining the physical position and orientation of sensors relative to each other. Intrinsic calibration: correcting a single sensor's own internal distortions (like lens distortion in a camera). Sensor fusion: combining data from multiple sensors into a single, more reliable estimate. Ground truth: an independently verified, trusted reference value used to check whether a sensor's readings are actually accurate.