In an effort to democratize the development of simultaneous localization and mapping (SLAM) technology, Google has open-sourced its Cartographer library for mapping environments in both 2D and 3D.
SLAM, of course, is an essential component of robot vacuums, drones, self-driving cars and other autonomous platforms. SLAM algorithms combine data from sensors to determine the position of each sensor and build a map of the surrounding environment. Take iRobot’s Roomba robot vacuums, for example. As a Roomba cleans your floor, it uses sensors and a camera to map your home.
“Our focus is on advancing and democratizing SLAM as a technology,” Google writes in a blog announcing the open-source Cartographer. “Currently, Cartographer is heavily focused on LIDAR SLAM. Through continued development and community contributions, we hope to add both support for more sensors and platforms as well as new features, such as lifelong mapping and localizing in a pre-existing map.”
Cartographer integrates with the Robot Operating System (ROS), and it’s already compatible with robotic platforms, including Toyota Human Support Robot (HSR), TurtleBots, PR2, and Revo LDS. James Kuffner, CTO at the Toyota Research Institute, detailed the HSR during his keynote at RoboBusiness 2016, saying it’ll be the standard for the July 2017 RoboCup in Nagoya, Japan. Here’s how Toyota is leveraging Cartographer on the HSR:
Google says it’s also releasing three years of LIDAR and IMU data that was collected using its 2D and 3D mapping backpack platforms during the development and testing of Cartographer.
If you’re interested in using the open-source Cartographer yourself, check it out on GitHUb here.
Google introduced Cartographer in 2014 as an indoor mapping tool in a backpack. A person wearing this backpack simply walks through any building and the floor plan is automatically generated in real time. The backpacker can also use a companion app to add points of interest while walking around the building.