Skip to Main Content
This paper develops a map generation method for AGVS by integrating multiple observation results considering the uncertainties in observation and motion. The map is incrementally generated by integrating each observation result into the latest map. Since the robot movement includes uncertainty, the AGV position is estimated before integration by matching the latest map and the new observation. Experiment results show that the map generation method can generate map for AGVS efficiently and precisely.