Development of a New Waypoint Navigation Algorithm Based on the Combination of Electronic Maps and Aerial Photographs

One of the primary tasks of an autonomous mobile robot is to navigate to its destination without colliding with obstacles. In general, it is advantageous to navigate the shortest path to the destination in terms of distance. However, for safe navigation in a realistic environment, waypoints should be assigned where the robot can travel on a known map and navigate to the destination. SLAM is an advanced algorithm that can estimate the mobile robot's self-localization and build accurate environmental maps. However, to apply the SLAM algorithm, the mobile robot must manually navigate to its destination to create an environment map and assign waypoints that may consume time. Since the SLAM algorithm is based on an incremental updating algorithm, depending on acquired environmental data, it might occasionally be failing both map building and self-localization. This paper describes a new waypoint assign system for the mobile robot based on inaccurate but available electronic maps and aerial photographs for unknown navigated environments. We generated a map suitable for mobile robot navigation and waypoints based on actual electronic maps and aerial photos to validate the proposed system. Based on the generated waypoints, we were able to perform waypoint navigation using an actual mobile robot.