Improving the robustness of Naïve Physics airflow mapping, using Bayesian reasoning on a multiple hypothesis tree

Previous work on odour localisation in enclosed environments, relying on an airflow model, has faced significant limitations due to the fact that large differences between airflow topologies are predicted for only small variations in a physical map. This is due to uncertainties in the map and approximations in the modelling process. Furthermore, there are uncertainties regarding the flow direction through inlet/outlet ducts. We have presented a method for dealing with these uncertainties, by generating multiple airflow hypotheses. As the robot performs odour localisation, airflow in the environment is measured and used to adjust the confidences of the hypotheses using Bayesian inference. The best hypothesis is then selected, which allows the completion of the localisation task. We have shown experimentally that this method is capable of improving the robustness of our method for odour localisation in the presence of uncertainties, where previously it was incapable. The results further demonstrate the usefulness of naive physics for practical robotics applications.

[1]  Bernard Meltzer,et al.  Analogical Representations of Naive Physics , 1989, Artif. Intell..

[2]  Ingemar J. Cox,et al.  Modeling a Dynamic Environment Using a Bayesian Multiple Hypothesis Approach , 1994, Artif. Intell..

[3]  Jonathan P. Bollback,et al.  Bayesian Inference of Phylogeny and Its Impact on Evolutionary Biology , 2001, Science.

[4]  Frank W. Grasso,et al.  Biomimetic robot lobster performs chemo-orientation in turbulence using a pair of spatially separated sensors: Progress and challenges , 2000, Robotics Auton. Syst..

[5]  R. Andrew Russell,et al.  To naively smell as no robot has smelt before , 2004, IEEE Conference on Robotics, Automation and Mechatronics, 2004..

[6]  Takamichi Nakamoto,et al.  Odor-source localization in the clean room by an autonomous mobile sensing system , 1996 .

[7]  W. Koch,et al.  On Bayesian MHT for well-separated targets in densely cluttered environment , 1995, Proceedings International Radar Conference.

[8]  Rodney M. Goodman,et al.  Swarm robotic odor localization , 2001, Proceedings 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems. Expanding the Societal Role of Robotics in the the Next Millennium (Cat. No.01CH37180).

[9]  Derek Dunn-Rankin,et al.  Measurement and prediction of indoor air flow in a model room , 2003 .

[10]  Rodney A. Brooks,et al.  A Robust Layered Control Syste For A Mobile Robot , 2022 .

[11]  George J. Klir,et al.  Fuzzy sets and fuzzy logic - theory and applications , 1995 .

[12]  Ingemar J. Cox,et al.  A Bayesian multiple-hypothesis approach to edge grouping and contour segmentation , 1993, International Journal of Computer Vision.

[13]  Keung-Chi Ng,et al.  Uncertainty management in expert systems , 1990, IEEE Expert.

[14]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[15]  Michael Negnevitsky,et al.  Artificial Intelligence: A Guide to Intelligent Systems , 2001 .

[16]  R. Andrew Russell,et al.  Bi-modal search using complementary sensing (olfaction/vision) for odour source localisation , 2006, Proceedings 2006 IEEE International Conference on Robotics and Automation, 2006. ICRA 2006..

[17]  R. Andrew Russell,et al.  Using naïve physics for odor localization in a cluttered indoor environment , 2006, Auton. Robots.

[18]  Gideon Kowadlo.  Advanced Airflow Modelling Using Naive Physics for Odour Localisation , 2005 .

[19]  R. Andrew Russell,et al.  Robot Odor Localization: A Taxonomy and Survey , 2008, Int. J. Robotics Res..

[20]  A. D. Young,et al.  An Introduction to Fluid Mechanics , 1968 .

[21]  Paul A. Boxer,et al.  Towards Learning Naive Physics by Visual Observation: Qualitative Spatial Representations , 2001, Australian Joint Conference on Artificial Intelligence.

[22]  Andreas Zell,et al.  Gas distribution in unventilated indoor environments inspected by a mobile robot , 2003 .

[23]  Rodney M. Goodman,et al.  Swarm robotic odor localization , 2001, Proceedings 2001 IEEE/RSJ International Conference on Intelligent Robots and Systems. Expanding the Societal Role of Robotics in the the Next Millennium (Cat. No.01CH37180).

[24]  Sebastian Thrun,et al.  Probabilistic robotics , 2002, CACM.

[25]  Jay A. Farrell,et al.  Biologically inspired chemical plume tracing on an autonomous underwater vehicle , 2004, 2004 IEEE International Conference on Systems, Man and Cybernetics (IEEE Cat. No.04CH37583).

[26]  G. Pulford.  Taxonomy of multiple target tracking methods , 2005 .

[27]  Greg Welch,et al.  An Introduction to Kalman Filter , 1995, SIGGRAPH 2001.

[28]  Patrick J. Hayes,et al.  The Naive Physics Manifesto , 1990, The Philosophy of Artificial Intelligence.

[29]  Hiroshi Ishida,et al.  Mobile robot navigation using vision and olfaction to search for a gas/odor source , 2006, Auton. Robots.

[30]  Gideon Kowadlo,et al.  Naive physics for effective odour localisation , 2003 .

[31]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[32]  Liliana Ardissono,et al.  Personalization in business-to-customer interaction , 2002, CACM.

[33]  Gideon Kowadlo.  Robot odour localisation in enclosed and cluttered environments using naïve physics , 2007 .

[34]  Mglc Marcel Loomans,et al.  The measurement and simulation of indoor air flow , 1998 .

[35]  R. Andrew Russell,et al.  Robotic location of underground chemical sources , 2004, Robotica.

[36]  R. Andrew Russell,et al.  A comparison of reactive robot chemotaxis algorithms , 2003, Robotics Auton. Syst..