Map drift is the gradual mismatch between a robot’s internal map and the real environment. It can occur when small errors in position estimates accumulate as the robot moves, causing mapped features to appear misplaced or the robot’s estimated location to become inaccurate.
In robotics and SLAM, drift matters because navigation depends on the map and position estimate staying aligned with reality. Wheel slip, noisy sensors, or limited processing can contribute. In an attack or defense context, unreliable maps can lead to unsafe movement or expose weaknesses in autonomous systems; testing across varied environments and checking estimates against reliable sensor data can help detect and reduce drift.