A New Strategy for Feature Initialization in Visual SLAM

This paper presents a Visual EKF-SLAM process using an original and very efficient strategy for initializing landmarks. Usually, with Cartesian coordinates, new points are created along the line-of-sight with a large variance. However, this type of initialization is subject to significant linearization issues making landmarks diverge from their real position. The immediate consequence is a failure of the Visual SLAM process. We propose here a new strategy that avoids or drastically limits the linearization errors. The first part of this strategy takes place during the tracking process where a coherent window is needed in order to successfully follow a point and make it converge. The second part concerns the update step. Due to linearization errors, a landmark in front of the observer can be updated behind it. We compute a corrective of the Kalman gain in order to preserve the integrity. We applied this strategy to real data illustrating its efficiency.

[1]  Javier Civera,et al.  Unified Inverse Depth Parametrization for Monocular SLAM , 2006, Robotics: Science and Systems.

[2]  Kostas E. Bekris,et al.  Evaluation of algorithms for bearing-only SLAM , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[3]  Carlo Tomasi,et al.  Good features to track , 1994, 1994 Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.

[4]  Michael Bosse,et al.  An Atlas framework for scalable mapping , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[5]  Tim Bailey.  Constrained initialisation for bearing-only SLAM , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[6]  Javier Civera,et al.  Inverse Depth to Depth Conversion for Monocular SLAM , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.

[7]  Romuald Aufrère,et al.  Improving results of rational non-linear observation functions using a Kalman filter correction , 2011, 14th International Conference on Information Fusion.

[8]  Hugh Durrant-Whyte,et al.  Simultaneous localization and mapping (SLAM): part II , 2006 .

[9]  Juan D. Tardós,et al.  Hierarchical SLAM: real-time accurate mapping of large environments , 2005, IEEE Transactions on Robotics.

[10]  Sebastian Thrun,et al.  FastSLAM: a factored solution to the simultaneous localization and mapping problem , 2002, AAAI/IAAI.

[11]  Hugh F. Durrant-Whyte,et al.  A solution to the simultaneous localization and map building (SLAM) problem , 2001, IEEE Trans. Robotics Autom..

[12]  Sebastian Thrun,et al.  Model based vehicle detection and tracking for autonomous urban driving , 2009, Auton. Robots.

[13]  Gamini Dissanayake,et al.  An efficient multiple hypothesis filter for bearing-only SLAM , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[14]  Walterio W. Mayol-Cuevas,et al.  Real-Time and Robust Monocular SLAM Using Predictive Multi-resolution Descriptors , 2006, ISVC.

[15]  Peter Cheeseman,et al.  A stochastic map for uncertain spatial relationships , 1988 .

[16]  Simon J. Julier,et al.  Avoiding negative depth in inverse depth bearing-only SLAM , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[17]  Javier Civera,et al.  Inverse Depth Parametrization for Monocular SLAM , 2008, IEEE Transactions on Robotics.

[18]  G. Klein,et al.  Parallel Tracking and Mapping for Small AR Workspaces , 2007, 2007 6th IEEE and ACM International Symposium on Mixed and Augmented Reality.

[19]  Andrew J. Davison,et al.  Real-time simultaneous localisation and mapping with a single camera , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[20]  Sebastian Thrun,et al.  FastSLAM: An Efficient Solution to the Simultaneous Localization And Mapping Problem with Unknown Data , 2004 .

[21]  Michel Devy,et al.  Undelayed initialization in bearing only SLAM , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[22]  Tom Drummond,et al.  Scalable Monocular SLAM , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[23]  Hugh F. Durrant-Whyte,et al.  Simultaneous localization and mapping: part I , 2006, IEEE Robotics & Automation Magazine.