Overview
Direct Answer
Simultaneous Localisation and Mapping (SLAM) is a computational technique enabling robots to construct a map of an unknown environment whilst concurrently determining their own position within that space. Unlike environments with pre-existing maps, SLAM operates in unmapped terrain by fusing sensor data—typically from cameras, lidar, or inertial measurement units—to solve the dual estimation problem in real time.
How It Works
SLAM algorithms use recursive filtering or graph-based optimisation to integrate incremental pose estimates with landmark observations. The system builds a probabilistic representation of environmental features whilst maintaining uncertainty estimates for both robot position and map structure. Loop closure detection refines the map by recognising previously visited locations, reducing drift accumulated from dead reckoning.
Why It Matters
Autonomous navigation in GPS-denied environments—indoors, underground, or underwater—depends on SLAM for operational autonomy without external infrastructure. This capability reduces deployment costs for mobile robots, autonomous vehicles, and drones whilst enabling real-time adaptation to dynamic or unmapped spaces.
Common Applications
Applications span autonomous vacuum cleaners navigating indoor spaces, wheeled robots in warehouses and manufacturing facilities, unmanned aerial vehicles surveying disaster sites, and subsea vehicles exploring underwater terrain. Research institutions employ SLAM in robotics laboratories for navigation benchmarking.
Key Considerations
Computational complexity scales with environment size and feature density, requiring careful trade-offs between real-time performance and map fidelity. Sensor noise, particularly in visually ambiguous or feature-poor environments, can accumulate significant localisation error without robust loop closure mechanisms.
Referenced By1 term mentions Simultaneous Localisation and Mapping
Other entries in the wiki whose definition references Simultaneous Localisation and Mapping — useful for understanding how this concept connects across Robotics & Automation and adjacent domains.
More in Robotics & Automation
Lights-Out Manufacturing
Industrial RoboticsA fully automated manufacturing process that requires no human presence on the factory floor.
Force Sensing
Industrial RoboticsThe capability of a robot to detect and measure physical forces applied to it during interaction with objects.
Robotic Process Mining
Software & AICombining process mining with robotic process automation to discover and automate business processes.
Robotic Vision
Software & AIThe use of cameras and computer vision algorithms to give robots the ability to see and interpret their environment.
Service Robot
Humanoid & SocialA robot that performs useful tasks for humans, operating semi or fully autonomously outside of industrial settings.
Humanoid Robot
Humanoid & SocialA robot designed to resemble the human body in shape and movement capabilities.
Digital Factory
Industrial RoboticsA comprehensive network of digital models and simulation tools replicating the entire production process.
Autonomous Mobile Robot
Autonomous SystemsA robot that can navigate and perform tasks in its environment without human guidance.