Impact of Landmark Parametrization on Monocular EKF-SLAM with Points and Lines
- 1.3k Downloads
This paper explores the impact that landmark parametrization has in the performance of monocular, EKF-based, 6-DOF simultaneous localization and mapping (SLAM) in the context of undelayed landmark initialization.
Undelayed initialization in monocular SLAM challenges EKF because of the combination of non-linearity with the large uncertainty associated with the unmeasured degrees of freedom. In the EKF context, the goal of a good landmark parametrization is to improve the model’s linearity as much as possible, improving the filter consistency, achieving robuster and more accurate localization and mapping.
This work compares the performances of eight different landmark parametrizations: three for points and five for straight lines. It highlights and justifies the keys for satisfactory operation: the use of parameters behaving proportionally to inverse-distance, and landmark anchoring. A unified EKF-SLAM framework is formulated as a benchmark for points and lines that is independent of the parametrization used. The paper also defines a generalized linearity index suited for the EKF, and uses it to compute and compare the degrees of linearity of each parametrization. Finally, all eight parametrizations are benchmarked employing analytical tools (the linearity index) and statistical tools (based on Monte Carlo error and consistency analyses), with simulations and real imagery data, using the standard and the robocentric EKF-SLAM formulations.
KeywordsMonocular vision Simultaneous localization and mapping Structure from motion Landmark parametrization Kalman filtering Benchmarking Linearity Consistency
Unable to display preview. Download preview PDF.
(MOV 4.62 MB)
(MOV 4.87 MB)
(MOV 4.80 MB)
(MOV 4.92 MB)
- Bailey, T. (2003). Constrained initialisation for bearing-only SLAM. In Int. conf. on robotics and automation (pp. 1966–1971). Google Scholar
- Bartoli, A., & Sturm, P. (2001). The 3D line motion matrix and alignment of line reconstructions. In IEEE computer society conference on computer vision and pattern recognition (Vol. 1, pp. 287–292). Google Scholar
- Berger, C., & Lacroix, S. (2010). DSeg: Détection directe de segments dans une image. In Reconnaissance des formes et intelligence artificielle. Google Scholar
- Bonarini, A., Burgard, W., Fontana, G., Matteucci, M., Sorrenti, D. G., & Tardos, J. D. (2006). RAWSEEDS: robotics advancement through web-publishing of sensorial and elaborated extensive data sets. In Proceedings of IROS’06 workshop on benchmarks in robotics research. Google Scholar
- Castellanos, J. A., Neira, J., & Tardós, J. D. (2004). Limits to the consistency of the EKF-based SLAM. In 5th IFAC symp. on intelligent autonomous vehicles, Lisboa, PT. Google Scholar
- Castellanos, J. A., Martinez-Cantin, R., Tardós, J. D., & Neira, J. (2007). Robocentric map joining: improving the consistency of EKF-SLAM. In Robotics and autonomous systems (Vol. 55, pp. 21–29). Google Scholar
- Chiuso, A., Favaro, P., Jin, H., & Soatto, S. (2002). Structure from motion causally integrated over time. In IEEE trans. on pattern analysis and machine intelligence (Vol. 24, pp. 523–535). Google Scholar
- Civera, J. (2009) Real-time EKF-based structure from motion. Ph.D. thesis, Universidad de Zaragoza. Google Scholar
- Civera, J., Grasa, O. G., Davison, A. J., & Montiel, J. M. M. (2009). 1-point RANSAC for EKF-based structure from motion. In IEEE/RSJ int. conf. on intelligent robots and systems. Google Scholar
- Davison, A. J. (2003). Real-time simultaneous localisation and mapping with a single camera. In Int. conf. on computer vision, Nice (Vol. 2, pp. 1403–1410). Google Scholar
- Eade, E., & Drummond, T. (2006a). Edge landmarks in monocular SLAM. In British machine vision conf., Edinburgh, Scotland. Google Scholar
- Eade, E., & Drummond, T. (2006b). Scalable monocular SLAM. IEEE International Conference on Computer Vision and Pattern Recognition, 1, 469–476. http://doi.ieeecomputersociety.org/10.1109/CVPR.2006.263. Google Scholar
- Eade, E., & Drummond, T. (2007). Monocular SLAM as a graph of coalesced observations. In IEEE int. conf. on computer vision. Google Scholar
- Engels, C., Stewénius, H., & Nistér, D. (2006). Bundle adjustment rules. In Photogrammetric computer vision. Google Scholar
- Gee, A. P., & Mayol, W. (2006). Real-time model-based SLAM using line segments. In LNCS proceedings of the 2nd international symposium on visual computing. Google Scholar
- Gee, A. P., Chekhlov, D., Calway, A., & Mayol-Cuevas, W. (2008). Discovering higher level structure in visual SLAM. In IEEE trans. on robotics special issue on visual SLAM (Vol. 24, pp. 980–990). Google Scholar
- Geeter, J. D., Brussel, H. V., Schutter, J. D., & Decréton, M. (1997). A smoothly constrained Kalman filter. In IEEE trans. on pattern analysis and machine intelligence (Vol. 24, pp. 1171–1177). Google Scholar
- Haner, S., & Heyden, A. (2010). On-line structure and motion estimation based on an novel parameterized extended Kalman filter. In Int. conf. on pattern recognition, Istambul, Turkey. Google Scholar
- Holmes, S. A., Klein, G., & Murray, D. W. (2008). A square root UKF for visual monoSLAM. In IEEE int. conf. on robotics and automation, Pasadena. Google Scholar
- Huang, S., & Dissanayake, G. (2007). Convergence and consistency analysis for extended Kalman filter based SLAM. In IEEE transactions on robotics (Vol. 23, pp. 1036–1049). Google Scholar
- Klein, G., & Murray, D. (2008). Improving the agility of keyframe-based SLAM. In 10th European conference on computer vision (pp. 802–815). Marseille. Google Scholar
- Kwok, N. M., & Dissanayake, G. (2003). Bearing-only SLAM in indoor environments using a modified particle filter. In Australasian conf. on robotics and automation (ACRA), Brisbane, Australia. Google Scholar
- Kwok, N. M., & Dissanayake, G. (2004). An efficient multiple hypothesis filter for bearing-only SLAM. In IEEE/RSJ int. conf. on intelligent robots and systems, Sendai, Japan. Google Scholar
- Lemaire, T., & Lacroix, S. (2007). Monocular-vision based SLAM using line segments. In IEEE int. conf. on robotics and automation (pp. 2791–2796). Rome, Italy. Google Scholar
- Lemaire, T., Lacroix, S., & Solà, J. (2005). A practical 3D bearing only SLAM algorithm. In IEEE/RSJ int. conf. on intelligent robots and systems, Edmonton, Canada. Google Scholar
- Lourakis, M., & Argyros, A. (2004). The design and implementation of a generic sparse bundle adjustment software package based on the levenberg-marquardt algorithm (Tech. Rep. 340). Institute of Computer Science—FORTH, Heraklion, Crete, Greece, available from http://www.ics.forth.gr/~lourakis/sba.
- Marzorati, D., Matteucci, M., Migliore, D., & Sorrenti, D. G. (2008). Monocular SLAM with inverse scaling parametrization. In Proc. of the British machine vision conference, Leeds. Google Scholar
- Montiel, J. M. M., Civera, J., & Davison, A. J. (2006). Unified inverse depth parametrization for monocular SLAM. In Robotics: science and systems, Philadelphia, USA. Google Scholar
- Piniés, P., Lupton, T., Sukkarieh, S., & Tardós, J. D. (2007). Inertial aiding of inverse depth SLAM using a monocular camera. In Int. conf. on robotics and automation. Google Scholar
- Smith, P., Reid, I., & Davison, A. J. (2006). Real-time monocular SLAM with straight lines. In British machine vision conf (Vol. 1, pp. 17–26). Google Scholar
- Solà, J. (2007). Towards visual localization, mapping and moving objects tracking by a mobile robot: a geometric and probabilistic approach. Ph.D. thesis, Institut National Polytechnique de Toulouse. Google Scholar
- Solà, J. (2010). Consistency of the monocular EKF-SLAM algorithm for 3 different landmark parametrizations. In IEEE int. conf. on robotics and automation, Anckorage, USA. Google Scholar
- Solà, J., Marquez, D., Codol, J. M., & Vidal-Calleja, T. (2009a). An EKF-SLAM toolbox for MATLAB. http://homepages.laas.fr/jsola/JoanSola/eng/toolbox.html.
- Solà, J., Vidal-Calleja, T., & Devy, M. (2009b). Undelayed initialization of line segments in monocular SLAM. In IEEE/RSJ int. conf. on intelligent robots and systems (pp. 1553–1558). Saint Louis, USA. Google Scholar
- Strasdat, H., Montiel, J. M. M., & Davison, A. J. (2010). Real-time monocular SLAM: Why filter? In Int. conf. on robotics and automation, Anckorage, USA. Google Scholar
- Sunderhauf, N., Lange, S., & Protzel, P. (2007). Using the unscented kalman filter in mono-SLAM with inverse depth parametrization for autonomous airship control. In IEEE int. workshop on safety, security and rescue robotics, Rome. Google Scholar