A Natural Navigation Method for Following Path Memories from 2D Maps

The navigation of mobile robots is an important issue in many areas such as industry, autonomous vehicles, and defense. Today, even the most known navigation systems, such as GPS, face natural and artificial problems in various operating environments. Solutions such as simultaneous localization and mapping, visual teach and repeat, and topological localization are suggested as a solution to these situations. In this study, a method is proposed for navigation to keep the bookmarks in the memory of living things and to reach the target using the key points. In this study, it was aimed to make use of the memories of a previously travelled path and then to follow the same path. In the presented method, 2D maps were collected with LIDAR while navigating the target path. Robot control was performed by using the previously prepared 2D map memory for matching and steering angle calculations. It was realized with the simulation that robots could follow the previously travelled path using this method. As a result, with the proposed method, mobile robots can reach the target with a low error rate even in multi-bend systems without error accumulation.