From Object Recognition to Activity Interpretation and Back, Based on Point Cloud Data

Semantic mapping of static environments has become a hot topic in robotics. The aim of the Mermaid project was to investigate the transfer of a sensor data interpretation approach for mapping to the problem of activity recognition in smart home applications such as elderly care. The basic structure of the semantic mapping approach, i.e., to assemble hypotheses of object aggregates in a closed-loop process of bottom-up raw data interpretation and top-down expectation generation from a domain ontology, can be extended to the temporal domain to include activity interpretation. This paper reports initial results, based on a study using point clouds from depth (RGB-D) sensor data.

[1]  Antony Galton,et al.  Towards an Integrated Logic of Space, Time and Motion , 1993, IJCAI.

[2]  Mehul Bhatt,et al.  Modelling Dynamic Spatial Systems in the Situation Calculus , 2008, Spatial Cogn. Comput..

[3]  J. Hertzberg,et al.  Matching CAD Object Models in Semantic Mapping , 2011 .

[4]  Joachim Hertzberg,et al.  Model-Based Object Recognition from 3D Laser Data , 2011, KI.

[5]  Reinhard Klein,et al.  Efficient RANSAC for Point‐Cloud Shape Detection , 2007, Comput. Graph. Forum.

[6]  Fulvio Mastrogiovanni,et al.  Proactive Assistance in Ecologies of Physically Embedded Intelligent Systems: A Constraint-Based Approach , 2011 .

[7]  Bernd Neumann,et al.  On scene interpretation with description logics , 2006, Image Vis. Comput..

[8]  Howard D. Wactlar,et al.  Dining activity analysis using a hidden Markov model , 2004, ICPR 2004.

[9]  William E. Lorensen,et al.  Marching cubes: A high resolution 3D surface construction algorithm , 1987, SIGGRAPH.

[10]  Daijin Kim,et al.  Simultaneous Gesture Segmentation and Recognition based on Forward Spotting Accumulative HMMs , 2006, 18th International Conference on Pattern Recognition (ICPR'06).

[11]  L. Benini,et al.  Activity recognition from on-body sensors by classifier fusion: sensor scalability and robustness , 2007, 2007 3rd International Conference on Intelligent Sensors, Sensor Networks and Information.

[12]  Hans W. Guesgen,et al.  Recognising Human Behaviour in a Spatio-Temporal Context , 2011 .

[13]  Stevan Harnad.  The Symbol Grounding Problem , 1999, ArXiv.

[14]  Nils J. Nilsson,et al.  Shakey the Robot , 1984 .

[15]  Alessandro Saffiotti,et al.  An introduction to the anchoring problem , 2003, Robotics Auton. Syst..

[16]  Hans W. Guesgen,et al.  Exploring the responsibilities of single-inhabitant Smart Homes with Use Cases , 2010, J. Ambient Intell. Smart Environ..

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

[18]  Reid G. Simmons,et al.  Robotic Systems Architectures and Programming , 2008, Springer Handbook of Robotics.

[19]  Tony DeRose,et al.  Surface reconstruction from unorganized points , 1992, SIGGRAPH.

[20]  Hans W. Guesgen,et al.  Behaviour Recognition from Sensory Streams in Smart Environments , 2009, Australasian Conference on Artificial Intelligence.

[21]  Kent Larson,et al.  Activity Recognition in the Home Using Simple and Ubiquitous Sensors , 2004, Pervasive.

[22]  Joachim Hertzberg,et al.  Mobile Roboter - Eine Einführung aus Sicht der Informatik , 2012, eXamen.press.

[23]  Wendong Xiao,et al.  Real-time Physical Activity classification and tracking using wearble sensors , 2007, 2007 6th International Conference on Information, Communications & Signal Processing.

[24]  Joachim Hertzberg,et al.  A Toolkit for Automatic Generation of Polygonal Maps - Las Vegas Reconstruction , 2012, ROBOTIK.

[25]  James F. Allen.  Maintaining knowledge about temporal intervals , 1983, CACM.

[26]  Jesse Hoey,et al.  Activity Recognition in Pervasive Intelligent Environments , 2011 .

[27]  Patrick Koopmann,et al.  SCENIOR: ontology-based interpretation of aircraft service activities , 2011 .