Unsupervised Learning for Subterranean Junction Recognition Based on 2D Point Cloud

This article proposes a novel unsupervised learning framework for detecting the number of tunnel junctions in subterranean environments based on acquired 2D point clouds. The implementation of the framework provides valuable information for high level mission planners to navigate an aerial platform in unknown areas or robot homing missions. The framework utilizes spectral clustering, which is capable of uncovering hidden structures from connected data points lying on non-linear manifolds. The spectral clustering algorithm computes a spectral embedding of the original 2D point cloud by utilizing the eigen decomposition of a matrix that is derived from the pairwise similarities of these points. We validate the developed framework using multiple data-sets, collected from multiple realistic simulations, as well as from real flights in underground environments, demonstrating the performance and merits of the proposed methodology.

[1]  Christos Papachristos,et al.  Autonomous Navigation and Mapping in Underground Mines Using Aerial Robots , 2019, 2019 IEEE Aerospace Conference.

[2]  Stephen Becker,et al.  Improved Fixed-Rank Nyström Approximation via QR Decomposition: Practical and Theoretical Aspects , 2017, Neurocomputing.

[3]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[4]  Hang Dong,et al.  Into Darkness: Visual Navigation Based on a Lidar-Intensity-Image Pipeline , 2013, ISRR.

[5]  Michael Himmelsbach,et al.  GIS-based topological robot localization through LIDAR crossroad detection , 2011, 2011 14th International IEEE Conference on Intelligent Transportation Systems (ITSC).

[6]  Ashok Samal,et al.  Semi-Automated Road Detection From High Resolution Satellite Images by Directional Morphological Enhancement and Segmentation Techniques , 2012, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[7]  ANNE MARSDEN,et al.  EIGENVALUES OF THE LAPLACIAN AND THEIR RELATIONSHIP TO THE CONNECTEDNESS , 2013 .

[8]  Fernando Santos Osório,et al.  Road junction detection from 3D point clouds , 2016, 2016 International Joint Conference on Neural Networks (IJCNN).

[9]  Johan A. K. Suykens,et al.  Fast kernel spectral clustering , 2017, Neurocomputing.

[10]  Stephen Becker,et al.  Preconditioned Data Sparsification for Big Data With Applications to PCA and K-Means , 2015, IEEE Transactions on Information Theory.

[11]  K. Madhava Krishna,et al.  Towards View-Invariant Intersection Recognition from Videos using Deep Network Ensembles , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[12]  John C. Platt,et al.  Learning Discriminative Projections for Text Similarity Measures , 2011, CoNLL.

[13]  K. Madhava Krishna,et al.  Have i reached the intersection: A deep learning-based approach for intersection detection from monocular cameras , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  S. N. Omkar,et al.  JuncNet: A Deep Neural Network for Road Junction Disambiguation for Autonomous Vehicles , 2018, ArXiv.

[15]  Paolo Gamba,et al.  Junction-aware extraction and regularization of urban road networks in high-resolution SAR images , 2006, IEEE Transactions on Geoscience and Remote Sensing.

[16]  G. Nikolakopoulos,et al.  Autonomous MAV Navigation in Underground Mines Using Darkness Contours Detection , 2019, ICVS.

[17]  Sina Sharif Mansouri,et al.  Vision-based MAV Navigation in Underground Mine Using Convolutional Neural Network , 2019, IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society.

[18]  Emil Fresk,et al.  Cooperative coverage path planning for visual inspection , 2018 .

[19]  David A. Clausi,et al.  Enabling scalable spectral clustering for image segmentation , 2010, Pattern Recognit..

[20]  Bo Wang,et al.  Visualization and analysis of single-cell RNA-seq data by kernel-based similarity learning , 2016, Nature Methods.

[21]  Ulrike von Luxburg,et al.  A tutorial on spectral clustering , 2007, Stat. Comput..

[22]  Sina Sharif Mansouri,et al.  Visual Subterranean Junction Recognition for MAVs based on Convolutional Neural Networks , 2019, IECON 2019 - 45th Annual Conference of the IEEE Industrial Electronics Society.

[23]  Andreas Loukas,et al.  Approximating Spectral Clustering via Sampling: a Review , 2019, Sampling Techniques for Supervised or Unsupervised Tasks.

[24]  Christoforos Kanellakis,et al.  Survey on Computer Vision for UAVs: Current Developments and Trends , 2017, Journal of Intelligent & Robotic Systems.

[25]  Chun Liu,et al.  A higher-order tensor voting-based approach for road junction detection and delineation from airborne LiDAR data , 2019, ISPRS Journal of Photogrammetry and Remote Sensing.

[26]  Darius Burschka,et al.  Toward a Fully Autonomous UAV: Research Platform for Indoor and Outdoor Urban Search and Rescue , 2012, IEEE Robotics & Automation Magazine.

[27]  Michael I. Jordan,et al.  On Spectral Clustering: Analysis and an algorithm , 2001, NIPS.

[28]  Pin-Yu Chen,et al.  Revisiting Spectral Graph Clustering with Generative Community Models , 2017, 2017 IEEE International Conference on Data Mining (ICDM).