Robot calibration of sensor poses and region based odometry using offline optimisation of map information

摘要: The calibration of mobile robot odometry and sensor extrinsic parameters can significantly improve the accuracy mapping localisation. This paper reports on simultaneous wheeled its onboard parameters. is achieved with offline optimisation map quality using a particle swarm algorithm. approach takes advantage state-of-the-art metric for 2D occupancy grid maps uses this measurement as fitness value optimization No ground truth required. Since depends floor surface type, improves previous efforts by employing novel colour classification system. Floor based Support Vector Machine that achieves greater than 98% precision recall values testing dataset consisting six different floors. We demonstrate benefit our proposed system in real world experiments. Furthermore, consistency method also validated experimentally data sets.