Towards Human Inspired Semantic SLAM

Robotic SLAM is attempting to learn robots what human beings do nearly effortlessly: to navigate in an unknown environment and to map it in the same time. In spite of huge advance in this area, nowadays SLAM solutions are not yet ready to enter the real world. In this paper, we observe the state of the art in existing SLAM techniques and identify semantic SLAM as one of prospective directions in robotic mapping research. We position our initial research into this field and propose a human inspired concept of SLAM based on understanding of the scene via its semantic analysis. First simulation results, using a virtual humanoid robot are presented to illustrate our approach.

[1]  James J. Little,et al.  Curious George: An attentive semantic robot , 2008, Robotics Auton. Syst..

[2]  Wolfram Burgard,et al.  Map building with mobile robots in dynamic environments , 2003, 2003 IEEE International Conference on Robotics and Automation (Cat. No.03CH37422).

[3]  Tom Duckett,et al.  Probabilistic Semantic Mapping with a Virtual Sensor for Building/Nature detection , 2007, 2007 International Symposium on Computational Intelligence in Robotics and Automation.

[4]  Roland Siegwart,et al.  Cognitive maps for mobile robots - an object based approach , 2007, Robotics Auton. Syst..

[5]  Cipriano Galindo,et al.  Multi-hierarchical semantic maps for mobile robotics , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[6]  Hugh Durrant-Whyte,et al.  Simultaneous Localisation and Mapping ( SLAM ) : Part I The Essential Algorithms , 2006 .

[7]  Joachim Hertzberg,et al.  Towards semantic maps for mobile robots , 2008, Robotics Auton. Syst..

[8]  Danica Kragic,et al.  Integrating Active Mobile Robot Object Recognition and SLAM in Natural Environments , 2006, 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[9]  David Fofi,et al.  Current state of the art of vision based SLAM , 2009, Electronic Imaging.