ROS-based 2-D Mapping Using Non-holonomic Differential Mobile Robot

Abstrak

This research proposes a 2-D mapping method by a mobile robot using LIDAR sensor. The mobile robot used is a non-holonomic type with a differential driver designed to operate in an indoor area. The robot applies an occupancy grid map method that uses a probability rule to handle the uncertainties of the sensor. The quality of 2-D occupied map relies on the accuracy of distance measurements by the LIDAR sensor and the accuracy of position estimation. Position estimation is obtained by using the 2-D LIDAR odometry which is based on the laser scan matching technique. This research uses simulation model which has characteristics like real nature. All the robotic software operations are managed by the Robot Operating System (ROS) as one of the most popular software frameworks currently used by robot researchers. The experimental results show that the robot can arrange a 2-D map well which is indicated by the similarity between the reference ground truth and the resulting 2-D map.