What is the main idea of the passage?

SLAM lets a robot estimate its location and build a map together, continually managing sensor uncertainty, drift, landmarks, and corrections. (The passage explains the circular dependency and the evidence-based processes used to handle it.)

Why is a recognized doorway useful after the robot returns?

It can connect the current position with an earlier one and help correct accumulated drift. (The loop-closure paragraph explains how a repeated place constrains and adjusts the earlier path.)

What does “drift” mean in the second paragraph?

gradual departure of an estimate from the robot's real path (The passage defines drift as small motion errors accumulating until estimated and real routes separate.)

Which device may estimate the robot's motion?

a camera (The second paragraph lists cameras among possible sensors.)

A Robot Maps While It Moves: Level 5 Reading Comprehension Questions

A robot enters an unfamiliar warehouse. To travel safely, it needs a map. To build that map, it must know where its sensor readings were taken. Yet finding its position also requires a map. This circular task is called simultaneous localization and mapping, or SLAM.

The robot estimates motion using wheel rotation, cameras, lidar, or other sensors. Each estimate carries error. A tiny mistake in turning can accumulate, causing the estimated path to drift away from the real one. Meanwhile, software identifies landmarks and records their probable locations. Neither pose nor map is treated as perfectly certain. Suppose the robot returns to a doorway it saw earlier. Recognizing that place creates a loop closure. The software can compare the new observation with the old one and adjust many earlier positions so the path joins consistently. A false match, however, can warp the map. Systems therefore weigh evidence and may wait for several agreeing features.

Warehouses add moving people, carts, blank walls, reflections, and changing stacks of boxes. A useful system must distinguish lasting landmarks from temporary obstacles. It may also combine a global map with rapid local decisions that avoid a cart crossing nearby.

SLAM does not give a robot one final, flawless picture. It maintains estimates and revises them as evidence arrives. The map helps locate the robot, and the location helps improve the map. Progress comes from managing this dependency, measuring uncertainty, and correcting accumulated error when the route supplies a trustworthy connection.

Comprehension questions