Machine Learning in Appearance-Based Robot Self-Localization

An appearance-based robot self-localization problem is considered in the machine learning framework. The appearance space is composed of all possible images, which can be captured by a robot's visual system under all robot localizations. Using recent manifold learning and deep learning techniques, we propose a new geometrically motivated solution based on training data consisting of a finite set of images captured in known locations of the robot. The solution includes estimation of the robot localization mapping from the appearance space to the robot localization space, as well as estimation of the inverse mapping for modeling visual image features. The latter allows solving the robot localization problem as the Kalman filtering problem.

[1]  Yoshua Bengio,et al.  How transferable are features in deep neural networks? , 2014, NIPS.

[2]  Alexander P. Kuleshov,et al.  Nonlinear multi-output regression on unknown input manifold , 2017, Annals of Mathematics and Artificial Intelligence.

[3]  M. Saito,et al.  Appearance based robot localization using regression models , 2006 .

[4]  Josef Pauli,et al.  Learning-Based Robot Vision , 2001, Lecture Notes in Computer Science.

[5]  Vincenzo Carnevale,et al.  Accurate Estimation of the Intrinsic Dimension Using Graph Distances: Unraveling the Geometric Complexity of Datasets , 2016, Scientific Reports.

[6]  Ben J. A. Kröse,et al.  Omnidirectional Vision for Appearance-Based Robot Localization , 2000, Sensor Based Intelligent Robots.

[7]  Evgeny Burnaev,et al.  Computationally efficient algorithm for Gaussian Process regression in case of structured samples , 2016, Computational Mathematics and Mathematical Physics.

[8]  Alexander P. Kuleshov,et al.  Statistical Learning via Manifold Learning , 2015, 2015 IEEE 14th International Conference on Machine Learning and Applications (ICMLA).

[9]  Clark F. Olson,et al.  Probabilistic self-localization for mobile robots , 2000, IEEE Trans. Robotics Autom..

[10]  James V. Candy,et al.  Model-Based Signal Processing , 1985 .

[11]  Francisco Bonin-Font,et al.  Visual Navigation for Mobile Robots: A Survey , 2008, J. Intell. Robotic Syst..

[12]  Ke Wang,et al.  Appearance-Based Map Learning for Mobile Robot by Using Generalized Regression Neural Network , 2007, ISNN.

[13]  Alexander P. Kuleshov,et al.  Tangent Bundle Manifold Learning via Grassmann&Stiefel Eigenmaps , 2012, ArXiv.

[14]  Huan N. Do,et al.  Appearance-Based Outdoor Localization Using Group LASSO Regression , 2015 .

[15]  J. Tenenbaum,et al.  A global geometric framework for nonlinear dimensionality reduction. , 2000, Science.

[16]  Ben J. A. Kröse,et al.  Robot environment modeling via principal component regression , 1999, Proceedings 1999 IEEE/RSJ International Conference on Intelligent Robots and Systems. Human and Environment Friendly Robots with High Intelligence and Emotional Quotients (Cat. No.99CH36289).

[17]  Evgeny Burnaev,et al.  Mobile Robot Localization via Machine Learning , 2017, MLDM.

[18]  Jongeun Choi,et al.  Appearance-based localization using Group LASSO regression with an indoor experiment , 2015, 2015 IEEE International Conference on Advanced Intelligent Mechatronics (AIM).

[19]  Alexander P. Kuleshov,et al.  Dimensionality Reduction in Statistical Learning , 2014, 2014 13th International Conference on Machine Learning and Applications.

[20]  Maxim Panov,et al.  Regression on the basis of nonstationary Gaussian processes with Bayesian regularization , 2016 .

[21]  Hobart R. Everett,et al.  Mobile robot positioning: Sensors and techniques , 1997, J. Field Robotics.

[22]  Alexey Zaytsev,et al.  Surrogate modeling of multifidelity data for large samples , 2015 .

[23]  John M. Lee.  Introduction to Smooth Manifolds , 2002 .

[24]  Alexander P. Kuleshov,et al.  Regression on High-Dimensional Inputs , 2016, 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW).

[25]  Alexander P. Kuleshov,et al.  Incremental Construction of Low-Dimensional Data Representations , 2016, ANNPR.

[26]  John Alan Fulcher,et al.  Artificial neural networks for pattern recognition , 1998 .

[27]  Jakub Šimánek.  Data fusion for localization using state estimation and machine learning , 2015 .

[28]  Bernt Schiele,et al.  Position estimation using principal components of range data , 1998, Robotics Auton. Syst..

[29]  R. Tibshirani,et al.  Regression shrinkage and selection via the lasso: a retrospective , 2011 .

[30]  Peter J. Bickel,et al.  Maximum Likelihood Estimation of Intrinsic Dimension , 2004, NIPS.

[31]  Antonino Staiano,et al.  Intrinsic dimension estimation: Advances and open problems , 2016, Inf. Sci..

[32]  Daniel D. Lee,et al.  Learning nonlinear appearance manifolds for robot localization , 2005, 2005 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[33]  David G. Lowe,et al.  Local and global localization for mobile robots using visual landmarks , 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).

[34]  Jake K. Aggarwal,et al.  Position estimation Techniques for an Autonomous Mobile robot - a Review , 1993, Handbook of Pattern Recognition and Computer Vision.

[35]  Alexander P. Kuleshov,et al.  Manifold Learning: Generalization Ability and Tangent Proximity , 2013, Int. J. Softw. Informatics.

[36]  Davide Scaramuzza,et al.  Benefit of large field-of-view cameras for visual odometry , 2016, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[37]  James L. Crowley,et al.  Continuity properties of the appearance manifold for mobile robot position estimation , 2001, Image Vis. Comput..