Experimental Analysis on Path Planning Strategy using Dijkstra Algorithm and Follow the Carrot Method for Indoor Mobile Robot Navigation

摘要

This research paper presents experimental analysis of global-local path planning strategy using Dijkstra algorithm and Follow the Carrot method implemented in differential drive mobile robot for indoor delivery mechanism. Experiment results show that Dijkstra algorithm performed an accurate, precise, and high stability of path planning strategy in supporting our indoor robot navigation. Experimental analysis has been performed to measure maximum positioning error, maneuverability of mobile robot, and accuracy of path tracking algorithm. Our research findings show that the more complicated the map in between the initial position and final position, the measured error rate has increased linearly. With a medium level of mobile robot's speed, which is set to 0.12 m/s, the measured maximum positioning error during path tracking process is 7.5 cm and the average positioning error is 2.5 cm. In terms of maneuverability, our mobile robot has demonstrated capability of coping with a sharp U-turn and eventually can correct its relative position to the targeted path.