Evolutionary robotics approach to odor source localization

An Evolutionary Robotics (ER) approach to the task of odor source localization is investigated. In particular, Continuous Time Recurrent Neural Networks (CTRNNs) are evolved for odor source localization in simulated turbulent odor plumes. In the experiments, the simulated robot is equipped with a single chemical sensor and a wind direction sensor. Three main contributions are made. First, it is shown that the ER approach can be successfully applied to odor source localization in both low-turbulent and high-turbulent conditions. Second, it is demonstrated that a small neural network is able to successfully perform all three sub-tasks of odor source localization: (i) finding the odor plume, (ii) moving toward the odor source, and (iii) identifying the odor source. Third, the analysis of the evolved behaviors reveals two novel odor source localization strategies. These strategies are successfully re-implemented as finite state machines, validating the insights from the analysis of the neural controllers.

[1]  H. Ishida,et al.  Chemical Sensing in Robotic Applications: A Review , 2012, IEEE Sensors Journal.

[2]  Xin Yao,et al.  Evolving artificial neural networks , 1999, Proc. IEEE.

[3]  Angelo Tartaglia,et al.  A relativistic navigation system for space , 2011 .

[4]  Dario Floreano,et al.  Neuroevolution: from architectures to learning , 2008, Evol. Intell..

[5]  Michael D. Smith,et al.  Strong Release of Methane on Mars in Northern Summer 2003 , 2009, Science.

[6]  Stefano Nolfi,et al.  Evolutionary Robotics: The Biology, Intelligence, and Technology of Self-Organizing Machines , 2000 .

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

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

[9]  Stefano Nolfi,et al.  Towards Pro-active Embodied Agents: On the Importance of Neural Mechanisms Suitable to Process Time Information , 2006 .

[10]  Dong Xiang,et al.  Metal Oxide Gas Sensors: Sensitivity and Influencing Factors , 2010, Sensors.

[11]  Dominique Martinez,et al.  Effectiveness and Robustness of Robot Infotaxis for Searching in Dilute Conditions , 2010, Front. Neurorobot..

[12]  Eduardo Serrano,et al.  Characterization of a clinical olfactory test with an artificial nose , 2011, Front. Neuroeng..

[13]  Richard S. Sutton,et al.  Reinforcement Learning: An Introduction , 2005, IEEE Transactions on Neural Networks.

[14]  KowadloGideon,et al.  Robot Odor Localization , 2008 .

[15]  Paolo Dario,et al.  SPIRAL: A novel biologically-inspired algorithm for gas/odor source localization in an indoor environment with no strong airflow , 2009, Robotics Auton. Syst..

[16]  Isao Shimoyama,et al.  Synthesis of pheromone-oriented emergent behavior of a silkworm moth , 1996, Proceedings of IEEE/RSJ International Conference on Intelligent Robots and Systems. IROS '96.

[17]  Alcherio Martinoli,et al.  A comparison of casting and spiraling algorithms for odor source localization in laminar flow , 2008, 2008 IEEE International Conference on Robotics and Automation.

[18]  Stanley J. Rosenschein,et al.  A dynamical systems perspective on agent-environment interaction , 1996 .

[19]  Richard S. Sutton,et al.  Introduction to Reinforcement Learning , 1998 .

[20]  Chris Melhuish,et al.  Some adaptive movements of animats with single symmetrical sensors , 1996 .

[21]  Massimo Vergassola,et al.  ‘Infotaxis’ as a strategy for searching without gradients , 2007, Nature.

[22]  J. Farrell,et al.  Chemical plume tracing experimental results with a REMUS AUV , 2003, Oceans 2003. Celebrating the Past ... Teaming Toward the Future (IEEE Cat. No.03CH37492).

[23]  David V. Thiel,et al.  A robotic system to locate hazardous chemical leaks , 1995, Proceedings of 1995 IEEE International Conference on Robotics and Automation.

[24]  Randall D. Beer,et al.  Evolving Dynamical Neural Networks for Adaptive Behavior , 1992, Adapt. Behav..

[25]  David C. Catling,et al.  Is there methane on Mars , 2010 .

[26]  Tobias Owen,et al.  Detection of methane in the martian atmosphere: evidence for life? , 2004 .

[27]  Paul F. M. J. Verschure,et al.  Moth-Like Chemo-Source Localization and Classification on an Indoor Autonomous Robot , 2011 .

[28]  Ian MacInnes,et al.  The Advantages of Evolving Perceptual Cues , 2006, Adapt. Behav..

[29]  Yuichi Nakamura,et al.  Approximation of dynamical systems by continuous time recurrent neural networks , 1993, Neural Networks.

[30]  Stefano Nolfi,et al.  Power and the limits of reactive agents , 2002, Neurocomputing.

[31]  Alcherio Martinoli,et al.  Theoretical analysis of three bio-inspired plume tracking algorithms , 2009, 2009 IEEE International Conference on Robotics and Automation.

[32]  J. H. Belanger,et al.  Biologically-inspired search algorithms for locating unseen odor sources , 1998, Proceedings of the 1998 IEEE International Symposium on Intelligent Control (ISIC) held jointly with IEEE International Symposium on Computational Intelligence in Robotics and Automation (CIRA) Intell.

[33]  Stefano Nolfi,et al.  Active Categorical Perception of Object Shapes in a Simulated Anthropomorphic Robotic Arm , 2010, IEEE Transactions on Evolutionary Computation.

[34]  Marco Giuranna,et al.  Detection of Methane in the Atmosphere of Mars , 2004, Science.

[35]  Pattie Maes,et al.  Some adaptive movements of animats with single symmetrical sensors , 1996 .

[36]  David J. Harvey,et al.  Comparing Insect-Inspired Chemical Plume Tracking Algorithms Using a Mobile Robot , 2008, IEEE Transactions on Robotics.

[37]  Jeffrey L. Elman,et al.  Finding Structure in Time , 1990, Cogn. Sci..

[38]  Shuo Pang,et al.  Chemical Plume Source Localization , 2006, IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics).

[39]  Paul A. Rhodes,et al.  Evolving a Neural Olfactorimotor System in Virtual and Real Olfactory Environments , 2012, Front. Neuroeng..

[40]  Dario Izzo,et al.  The Generalized Island Model , 2012, Parallel Architectures and Bioinspired Algorithms.

[41]  Randall D. Beer,et al.  The Dynamics of Active Categorical Perception in an Evolved Model Agent , 2003, Adapt. Behav..

[42]  E. Postma,et al.  Evolutionary Learning Outperforms Reinforcement Learning on Non-Markovian Tasks , 2005 .

[43]  Rodney M. Goodman,et al.  Distributed odor source localization , 2002 .

[44]  M EM,et al.  Acta Futura Toward Neuromorphic Odor Tracking : Perspectives for space exploration , 2011 .

[45]  Paul R. Cohen,et al.  Empirical methods for artificial intelligence , 1995, IEEE Expert.

[46]  S. Lockery,et al.  Evolution and Analysis of Minimal Neural Circuits for Klinotaxis in Caenorhabditis elegans , 2010, The Journal of Neuroscience.

[47]  Shimon Whiteson,et al.  Evolutionary Computation for Reinforcement Learning , 2012, Reinforcement Learning.

[48]  Lino Marques,et al.  Olfaction-based mobile robot navigation , 2002 .

[49]  Eduardo Izquierdo-Torres,et al.  Analysis of a Dynamical Recurrent Neural Network Evolved for Two Qualitatively Different Tasks: Walking and Chemotaxis , 2008, ALIFE.

[50]  Franck Lefèvre,et al.  Observed variations of methane on Mars unexplained by known atmospheric chemistry and physics , 2009, Nature.

[51]  Isao Shimoyama,et al.  A Pheromone-Guided Mobile Robot that Behaves like a Silkworm Moth with Living Antennae as Pheromone Sensors , 1998, Int. J. Robotics Res..

[52]  L. Marques,et al.  Electronic nose-based odour source localization , 2000, 6th International Workshop on Advanced Motion Control. Proceedings (Cat. No.00TH8494).

[54]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .