Improving Localization by Learning Pole-Like Landmarks Using a Semi-supervised Approach
The aim of this paper is to contribute with an object-based learning and selection methods to improve localization and mapping techniques. The methods use 3D-LiDAR data which is suitable for autonomous driving systems operating in urban environments. The objects of interest to be served as landmarks are pole-like objects which are naturally present in the environment. To detect and recognize pole-like objects in 3D-LiDAR data, a semi-supervised iterative label propagation method has been developed. Additionally, a selection method is proposed for selection the best poles to be used in the localization loop. The LiDAR localization and mapping system is validated using data from the KITTI database. Reported results show that by considering the occurrence of pole-like objects over time leads to an improvement on both the learning model and the localization.
KeywordsLiDAR odometry Semi-supervised learning Incremental label propagation SLAM Pole-based localization
This work was supported partially by the project MATIS (CENTRO-01-0145-FEDER-000014) co-financed by the European Regional Development Fund (FEDER) through of the Centro Regional Operacional Program (CENTRO2020), Portugal. It was also partially supported by the University of Coimbra, Institute of Systems and Robotics (ISR-UC) and FCT (Portuguese Science Foundation) through grant UID/EEA/00048/2019.
- 3.Chiotellis, I., Zimmermann, F., Cremers, D., Triebel, R.: Incremental semi-supervised learning from streams for object classification. In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 5743–5749 (2018)Google Scholar
- 4.Ester, M., Kriegel, H.P., Sander, J., Xu, X., et al.: A density-based algorithm for discovering clusters in large spatial databases with noise. In: KDD, pp. 226–231 (1996)Google Scholar
- 5.Garrote, L., Torres, M., Barros, T., Perdiz, J., Premebida, C., Nunes, U.J.: Mobile robot localization with reinforcement learning map update decision aided by an absolute indoor positioning system. In: 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (2019)Google Scholar
- 7.Geiger, A., Lenz, P., Stiller, C., Urtasun, R.: Vision meets robotics: the KITTI dataset. Int. J. Robot. Res. (IJRR) 14(3), 195–210 (2013)Google Scholar
- 8.Jacobson, A., Zeng, F., Smith, D., Boswell, N., Peynot, T., Milford, M.: Semi-supervised SLAM: leveraging low-cost sensors on underground autonomous vehicles for position tracking. In: 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 3970–3977 (2018)Google Scholar
- 9.Kampker, A., Hatzenbuehler, J., Klein, L., Sefati, M., Kreiskoether, K.D., Gert, D.: Concept study for vehicle self-localization using neural networks for detection of pole-like landmarks. In: International Conference on Intelligent Autonomous Systems, pp. 689–705. Springer (2018)Google Scholar
- 10.Lam, J., Kusevic, K., Mrstik, P., Harrap, R., Greenspan, M.: Urban scene extraction from mobile ground based LIDAR data. In: Proceedings of 3DPVT, pp. 1–8 (2010)Google Scholar
- 13.Schaefer, A., Büscher, D., Vertens, J., Luft, L., Burgard, W.: Long-term urban vehicle localization using pole landmarks extracted from 3-D lidar scans (2019)Google Scholar
- 15.Sefati, M., Daum, M., Sondermann, B., Kreisköther, K.D., Kampker, A.: Improving vehicle localization using semantic and pole-like landmarks. In: 2017 IEEE Intelligent Vehicles Symposium (IV), pp. 13–19 (2017)Google Scholar
- 16.Teichman, A., Thrun, S.: Group induction. In: 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 2757–2763 (2013)Google Scholar
- 17.Weng, L., Yang, M., Guo, L., Wang, B., Wang, C.: Pole-based real-time localization for autonomous driving in congested urban scenarios. In: 2018 IEEE International Conference on Real-time Computing and Robotics (RCAR), pp. 96–101 (2018)Google Scholar
- 19.Zhu, X., Ghahramani, Z.: Learning from labeled and unlabeled data with label propagation. Tech. rep. CMU-CALD-02-107, Carnegie Mellon University (2002)Google Scholar