Visual SLAM and Moving-object Detection for a Small-size Humanoid Robot

In the paper, a novel moving object detection (MOD) algorithm is developed and integrated with robot visual Simultaneous Localization and Mapping (vSLAM). The moving object is assumed to be a rigid body and its coordinate system in space is represented by a position vector and a rotation matrix. The MOD algorithm is composed of detection of image features, initialization of image features, and calculation of object coordinates. Experimentation is implemented on a small-size humanoid robot and the results show that the performance of the proposed algorithm is efficient for robot visual SLAM and moving object detection.

[1]  Luc Van Gool,et al.  SURF: Speeded Up Robust Features , 2006, ECCV.

[2]  Tony Lindeberg,et al.  Feature Detection with Automatic Scale Selection , 1998, International Journal of Computer Vision.

[3]  Bruno Siciliano,et al.  Modelling and Control of Robot Manipulators , 1997, Advanced Textbooks in Control and Signal Processing.

[4]  Olivier Stasse,et al.  MonoSLAM: Real-Time Single Camera SLAM , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[5]  Hugh F. Durrant-Whyte,et al.  Simultaneous Localization, Mapping and Moving Object Tracking , 2007, Int. J. Robotics Res..

[6]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

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

[8]  Peter I. Corke,et al.  A tutorial on visual servo control , 1996, IEEE Trans. Robotics Autom..

[9]  Luc Van Gool,et al.  Speeded-Up Robust Features (SURF) , 2008, Comput. Vis. Image Underst..

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