Simultaneous localization and mapping (also known as SLAM) is an algorithm that allows autonomous mobile robots or vehicles to construct a map of their surroundings and determine their location in that environment.
The acronym SLAM first appeared in 1995, while the algorithms that solved this mathematical problem started seeing real-world use in the early 21st century.
Without these algorithms, robots would not be able to build maps and keep track of their location. Without SLAM, robots can not be considered autonomous.
Simultaneous localization and mapping can work based on various sensor inputs such as:
LiDAR
Stereo cameras
Sonar sensors
When thinking about SLAM, people often think that it is only used in the initial mapping of an area. However, advanced SLAM algorithms can continuously update maps. This is especially useful for sites that the robot already knows but that change frequently, such as a warehouse.
Whether it’s due to pallets being moved around, corridors being blocked, or random obstacles, SLAM will allow the autonomous robots to map all of these and navigate accordingly.