Localization Design and Implementation Using Combined Rotary Encoder, IMU, and ROS on Delivery Service Robots in Building Area

Abstract

Developing delivery service robots requires an accurate localization system to send delivery objects to the right and appropriate locations. Localization of the robot is needed to determine its position and displacement at the intended point. To perform localization, precise sensors and appropriate algorithms are required to perform localization. Using localization only determines precision using one rotary encoder sensor, it is tough to get localization precision. Therefore, it is necessary to combine sensors and get the right algorithm to get localization precision, using a combination of Rotary Encoder and IMU sensors using EKF algorithms and robot operating system makes it easier in communicating, build a robot, and configure the setup. From the result of the tests carried out when moving linearly using a Rotary Encoder sensor, the average error result is obtained as 0.89602 m, and the result when using a combined sensor, between the rotary encoder and IMU, the average error results obtained are 0.85951 m. By using a combined sensor, it has been proven that the combination of these sensors produces a smaller distance error, so that when using a combined sensor the robot's precision level when moving is very small.