What Is Simultaneous Localization and Mapping?

Written by Coursera Staff • Updated on

Simultaneous localization and mapping is key for the future of autonomous robotics. Learn about simultaneous localization and mapping, including what simultaneous localization and mapping is and simultaneous localization and mapping applications.

[Featured Image] Three engineers discuss simultaneous localization and mapping while looking at a drone.

The next time you turn on a robot vacuum, watch its movements while cleaning a room. It might surprise you that the vacuum doesn’t bounce randomly around the space but moves efficiently to cover and clean the entire area. Simultaneous localization and mapping is autonomously navigating a room, road, or other dynamic area. 

Learn more about simultaneous localization and mapping, how this technology functions, simultaneous localization and mapping applications, and how to start a career in this field.

What is simultaneous localization and mapping?

Simultaneous localization and mapping (SLAM) is the ability of a robot to actively generate the map of an area while also keeping track of its own location. Localization refers to the robot’s ability to determine its own place within an area, and mapping refers to its process of building a map of the room or space it’s navigating. SLAM is key for robots to move successfully around unknown or changing environments without getting lost or stuck.

What is simultaneous localization and mapping used for?

Simultaneous localization and mapping applications range across many industries that require robotics to have the ability to navigate themselves. Some of these applications include:

Self-driving cars

Due to the rapidly changing nature of the road, particularly in the cities, self-driving cars cannot rely on GPS for orientation alone. With SLAM, self-driving cars utilize a wide range of sensors that provide real-time information about the car’s location while also supplying details about the location of other cars, pedestrians, traffic signs, and obstacles. 

Robot vacuums

If you have a robot vacuum, it uses SLAM to efficiently reorient itself to a space, such as a living room, that has the potential to change its layout between uses. The robot uses fixed points, such as the walls, to navigate the room while quickly adapting to differences, like if you moved a coffee table or lamp. SLAM allows your robot vacuum to update its map instantaneously without needing to redraw it every time it’s turned on.

Autonomous transport systems

Autonomous transport systems use SLAM to navigate spaces like warehouses or hospitals successfully. The robots work together to complete repetitive tasks, such as transferring packages, without the need for human intervention. This allows the employees to focus on more important tasks like solving problems and designing future plans. Additionally, autonomous transport systems also have the ability to interact with elevators and doors.

How does simultaneous localization and mapping work?

Simultaneous localization and mapping works by using two components—sensor-signal processing and pose-graph optimization—to orient and move a robot around a space successfully. 

The sensor-signal processing component successfully occurs through either visual or LiDAR SLAM. Visual SLAM uses cameras to examine a space and create a map to follow. The cameras allow the robot to capture high-resolution images to help it understand the composition of the space in which it’s working. 

LiDAR SLAM, or light detection and ranging (LiDAR), uses a laser or distance sensor to map space. LiDAR tracks the depth and geometry of the objects around it, which provides a very accurate and efficient map. Additionally, recent innovations in this technology, such as the creation of LiDAR feature detectors, will allow this system to map almost any environment.

After the robot collects the information it gathers in its sensors, pose-graph optimization occurs, which means it uses that information to create the map of the area it is attempting to navigate.

Who uses simultaneous localization and mapping?

Industries that employ autonomous robotics use simultaneous localization and mapping. From Mars to your own home, some examples include field robots, such as landers and rovers, robot vacuums, self-driving cars, and autonomous transport systems in factories and warehouses. Factories and warehouses utilize these autonomous transport systems, which apply simultaneous localization and mapping to deliver packages or sort shelves, park a self-driving car in a lot, or navigate a pilotless drone. Finally, while still in the early stages of its development, visual SLAM could be the key for augmented reality applications.

Pros and cons of using simultaneous localization and mapping

The pros and cons of using simultaneous localization and mapping involve the accuracy of the data the robot accumulates. Some pros include the increased efficiency of a robot as it completes its tasks. Another benefit is that these robots create reliable maps that allow them to avoid obstacles.

SLAM is especially helpful in environments where it’s impossible to map out the terrain before sending in a robot. Some examples of this type of location include mineshafts, dangerous areas such as volcanic tunnels, or the surface of Mars.

Some cons of SLAM include the risk of accumulating localization errors, which results in inaccurate maps over time, or the loss of localization so that the robot loses its own position or orientation on the map. Countermeasures, such as creating landmarks, help to reduce the risk of these issues and create more efficient maps over time. It’s also more difficult to apply SLAM technology in dynamic situations, such as busy roads, versus static spaces like a living room.

How to start a career in simultaneous location and mapping

If you’re interested in a career that involves simultaneous location and mapping, a position in robotics is a good choice. For example, a robotics engineer is responsible for using resources and tools like SLAM to create machines that take over menial tasks from human professionals. You research, design, build, implement, and maintain robotics systems in this role. According to Glassdoor, if you choose to work as a robotics engineer, you can expect to earn an average salary of $95,718 [1].

To become a robotics engineer, you’ll need a four-year or advanced degree in a field like computer science or engineering. Along with a degree, to perform in this position, you will need a working knowledge of multiple programming languages and a passion for problem-solving and computers. Then, it is smart to select a specific area of robotics to focus your efforts. A few of these to choose from are software, hardware, or connectivity. Finally, once you’re ready to work, you must become licensed, and since each state has different requirements, make sure to do your research.

Getting started with Coursera

Sharpen your robotics skills and learn about simultaneous localization and mapping with courses and degrees offered on Coursera. With options such as the IBM Machine Learning Professional Certificate, you’ll discover the foundational skills and knowledge needed for success in a robotics engineering career and what it’s like to use tools like simultaneous localization and mapping. Introduce yourself to the exciting and dynamic world of robotics while developing important skills on Coursera.

Article sources

  1. Glassdoor. “Robotics Engineer Overview, https://www.glassdoor.com/Career/robotics-engineer-career_KO0,17.htm#:~:text=%2495%2C451,Base%20Pay.” Accessed March 22, 2024.

Keep reading

Updated on
Written by:

Editorial Team

Coursera’s editorial team is comprised of highly experienced professional editors, writers, and fact...

This content has been made available for informational purposes only. Learners are advised to conduct additional research to ensure that courses and other credentials pursued meet their personal, professional, and financial goals.