Artificial Neural Networks and Machine Learning – ICANN 2016

This paper presents an artificial neural network-based control architecture allowing autonomous mobile robot indoor navigation by emulating the cognition process of a human brain when navigating in an unknown environment. The proposed architecture is based on a simultaneous top-down and bottom up approach, which combines the a priori knowledge of the environment gathered from a previously examined floor plan with the visual information acquired in real time. Thus, in order to take the right decision during navigation, the robot is able to process both set of information, compare them in real time and react accordingly. The architecture is composed of two modules: a) A deliberative module, corresponding to the processing chain in charge of extracting a sequence of navigation signs expected to be found in the environment, generating an optimal path plan to reach the goal,computing and memorizing the sequence of signs [1]. The path planning stage allowing the computation of the sign sequence is based on a neural implementation of the resistive grid. b) A reactive module, integrating the said sequence information in order to use it to control online navigation and learning sensory-motor associations. It follows a perception-action mechanism that constantly evolves because of the dynamic interaction between the robot and its environment. It is composed of three layers: one layer using a cognitive mechanism and the other two using a reflex mechanism. Experimental results obtained from the physical implementation of the architecture in an indoor environment show the feasibility of this approach.

[1]  Pavel Yu. Chebotarev,et al.  The Matrix-Forest Theorem and Measuring Relations in Small Social Groups , 2006, ArXiv.

[2]  Ali S. Hadi,et al.  Finding Groups in Data: An Introduction to Chster Analysis , 1991 .

[3]  Ulrike von Luxburg,et al.  Hitting and commute times in large random neighborhood graphs , 2014, J. Mach. Learn. Res..

[4]  F. Radicchi,et al.  Benchmark graphs for testing community detection algorithms. , 2008, Physical review. E, Statistical, nonlinear, and soft matter physics.

[5]  W. W. Daniel,et al.  Applied Nonparametric Statistics , 1978 .

[6]  Leo Grady,et al.  The Graph Analysis Toolbox: Image Processing on Arbitrary Graphs , 2003 .

[7]  François Fouss,et al.  Randomized Shortest-Path Problems: Two Related Models , 2009, Neural Computation.

[8]  M. Newman,et al.  Finding community structure in networks using the eigenvectors of matrices. , 2006, Physical review. E, Statistical, nonlinear, and soft matter physics.

[9]  François Bavaud,et al.  Interpolating between Random Walks and Shortest Paths: A Path Functional Approach , 2012, SocInfo.

[10]  K. Aihara,et al.  Chaotic neural networks , 1990 .

[11]  Luh Yen,et al.  A family of dissimilarity measures between nodes generalizing both the shortest-path and the commute-time distances , 2008, KDD.

[12]  Richard O. Duda,et al.  Pattern classification and scene analysis , 1974, A Wiley-Interscience publication.

[13]  Mark Newman,et al.  Networks: An Introduction , 2010 .

[14]  Janez Demsar,et al.  Statistical Comparisons of Classifiers over Multiple Data Sets , 2006, J. Mach. Learn. Res..

[15]  Jean-Loup Guillaume,et al.  Fast unfolding of communities in large networks , 2008, 0803.0476.

[16]  Marco Saerens,et al.  Developments in the theory of randomized shortest paths with a comparison of graph node distances , 2012, 1212.1666.

[17]  Guy Marchal,et al.  Automated multi-modality image registration based on information theory , 1995 .

[18]  Yi Sun,et al.  From random walks to distances on unweighted graphs , 2015, NIPS.

[19]  Pavel Yu. Chebotarev,et al.  A class of graph-geodetic distances generalizing the shortest-path and the resistance distances , 2008, Discret. Appl. Math..

[20]  M E J Newman,et al.  Modularity and community structure in networks. , 2006, Proceedings of the National Academy of Sciences of the United States of America.

[21]  U. V. Luxburg,et al.  Getting lost in space: large sample analysis of the commute distance , 2010, NIPS 2010.

[22]  François Fouss,et al.  Graph Nodes Clustering Based on the Commute-Time Kernel , 2007, PAKDD.

[23]  Ronald L. Rivest,et al.  Introduction to Algorithms, third edition , 2009 .

[24]  François Fouss,et al.  Graph nodes clustering with the sigmoid commute-time kernel: A comparative study , 2009, Data Knowl. Eng..

[25]  Bernice W. Polemis.  Nonparametric Statistics for the Behavioral Sciences , 1959 .

[26]  Alexander J. Smola,et al.  Learning with kernels , 1998 .

[27]  P. Groenen,et al.  Modern multidimensional scaling , 1996 .

[28]  BART KOSKO,et al.  Bidirectional associative memories , 1988, IEEE Trans. Syst. Man Cybern..

[29]  L. Hubert,et al.  Comparing partitions , 1985 .

[30]  M E J Newman,et al.  Community structure in social and biological networks , 2001, Proceedings of the National Academy of Sciences of the United States of America.

[31]  Celeux Gilles,et al.  Classification automatique des données , 2017 .

[32]  Pavel Yu. Chebotarev,et al.  The graph bottleneck identity , 2010, Adv. Appl. Math..

[33]  François Fouss,et al.  The Sum-over-Forests Density Index: Identifying Dense Regions in a Graph , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[34]  M E J Newman,et al.  Finding and evaluating community structure in networks. , 2003, Physical review. E, Statistical, nonlinear, and soft matter physics.

[35]  Kevin Françoisse,et al.  A bag-of-paths framework for network data analysis , 2013, Neural Networks.