The line follower-based AGV navigation system used at PT Akebono has limited flexibility when dealing with layout changes and path obstacles. The physical track used by the line follower system requires manual adjustment, which is time-consuming and costly, and causes operational delays when the robot encounters unavoidable obstructions. This research aims to develop a navigation system based on Simultaneous Localization and Mapping (SLAM) as a more flexible alternative. The methodology used includes experimental testing and quantitative data collection through speed tests, accuracy evaluations, load simulations, turning path changes, and user experience questionnaires. The AGV prototype was modified with LIDAR sensors, IMU, Mini PC, and ROS as the main control platform. The results show that the SLAM system can operate without physical tracks, perform automatic rerouting when encountering obstacles, and significantly reduce the time needed to change paths. The average reroute time for
SLAM was only 34-37 seconds, compared to 277-298 seconds for the line follower. In addition, SLAM received positive feedback from users, with an average satisfaction score of 4.63 out of 5. The system effectively enhances the flexibility
and operational efficiency of AGVs in dynamic production environments.
Keywords: navigation system, SLAM, AGV, operational flexibility, line follower