One of the main problems of the explosive growth of robots is that the code cannot be reused on different robot platforms. However, the hardware abstraction layer in ROS and its message service allow the creation of new code that can be used in many different robot platforms. Moreover, ROS provides a set of stable robot software packages, and the recognized SLAM evaluation methods all rely on the standard data sets available in the robot community. ?

All SLAM results take the occupied grid as the final output, and the performance index of map similarity is used for analysis. The emphasis is on the map quality, not the attitude estimation error, because the map output is greatly affected by the positioning problem.

Typical application fields of SLAM:

Map modeling. SLAM can assist robots to perform tasks such as path planning, autonomous exploration and navigation. Cobos, Tammy and the latest Lanbao sweeping robot in China, through the combination of SLAM algorithm and lidar or camera, can make the sweeper draw indoor maps efficiently, intelligently analyze and plan the sweeping environment, and thus successfully enter the array of intelligent navigation.

SLAMTEC is the main provider of this technology in China. SLAMTEC is named after SLAM's homonym, and its main business is to study the solution of autonomous positioning and navigation for service robots.