SLAM

SLAM (Simultaneous Localization and Mapping) is the family of methods a robot uses to build a map of an environment it has never seen while estimating its own position within that map. The two problems depend on each other: an accurate position needs a map, and an accurate map needs known positions, so SLAM solves them together. Common variants use lidar scans (lidar SLAM), camera images (visual SLAM), or both, often fused with wheel odometry and an inertial measurement unit.

The distinction matters because SLAM is what lets a mobile robot work without fixed infrastructure such as floor tape, magnets or reflectors. It is standard in autonomous mobile robots, robot vacuums, inspection quadrupeds and many humanoids, and it is why those machines can be deployed into a new building in days rather than weeks.