Training a terrain traversability classifier for a planetary rover through simulation

A classifier training methodology is presented for Kapvik, a micro-rover prototype. A simulated light detection and ranging scan is divided into a grid, with each cell having a variety of characteristics (such as number of points, point variance and mean height) which act as inputs to classification algorithms. The training step avoids the need for time-consuming and error-prone manual classification through the use of a simulation that provides training inputs and target outputs. This simulation generates various terrains that could be encountered by a planetary rover, including untraversable ones, in a random fashion. A sensor model for a three-dimensional light detection and ranging is used with ray tracing to generate realistic noisy three-dimensional point clouds where all points that belong to untraversable terrain are labelled explicitly. A neural network classifier and its training algorithm are presented, and the results of its output as well as other popular classifiers show high accuracy on test data sets after training. The network is then tested on outdoor data to confirm it can accurately classify real-world light detection and ranging data. The results show the network is able to identify terrain correctly, falsely classifying just 4.74% of untraversable terrain.

[1]  Toyomi Fujita,et al.  3D terrain measurement system with movable laser range finder , 2009, 2009 IEEE International Workshop on Safety, Security & Rescue Robotics (SSRR 2009).

[2]  Roberto Manduchi,et al.  Supervised Parametric Classification of Aerial LiDAR Data , 2004, 2004 Conference on Computer Vision and Pattern Recognition Workshop.

[3]  Mahmoud Tarokh,et al.  Kinematics modeling and analyses of articulated rovers , 2005, IEEE Transactions on Robotics.

[4]  Sebastian Scherer,et al.  VoxNet: A 3D Convolutional Neural Network for real-time object recognition , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[5]  Tom Duckett,et al.  Scan registration for autonomous mining vehicles using 3D‐NDT , 2007, J. Field Robotics.

[6]  Joachim Hertzberg,et al.  Automatic Classification of Objects in 3D Laser Range Scans , 2003 .

[7]  Sebastian Thrun,et al.  FastSLAM 2.0: an improved particle filtering algorithm for simultaneous localization and mapping that provably converges , 2003, IJCAI 2003.

[8]  Simon Haykin,et al.  Neural Networks and Learning Machines , 2010 .

[9]  Martin Buss,et al.  Comparison of surface normal estimation methods for range sensing applications , 2009, 2009 IEEE International Conference on Robotics and Automation.

[10]  Alex Ellery,et al.  Terrain Response Estimation Using an Instrumented Rocker-Bogie Mobility System , 2013, IEEE Transactions on Robotics.

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

[12]  G. Lewicki,et al.  Approximation by Superpositions of a Sigmoidal Function , 2003 .

[13]  Vivek Verma,et al.  3D Building Detection and Modeling from Aerial LIDAR Data , 2006, 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06).

[14]  Eric Haines.  FAST RAY – CONVEX POLYHEDRON INTERSECTION , 1991 .

[15]  Roberto Manduchi,et al.  Fast and reliable obstacle detection and segmentation for cross-country navigation , 2002, Intelligent Vehicle Symposium, 2002. IEEE.

[16]  G. Walton,et al.  An approach for automated lithological classification of point clouds , 2016 .

[17]  Fernando Santos Osório,et al.  Terrain mapping and classification using neural networks , 2009, ICHIT '09.

[18]  Lee A. Feldkamp,et al.  Decoupled extended Kalman filter training of feedforward layered networks , 1991, IJCNN-91-Seattle International Joint Conference on Neural Networks.

[19]  E. Stoll,et al.  Monte Carlo Calculation for Electromagnetic-Wave Scattering from Random Rough Surfaces , 1984 .

[20]  Dietrich Paulus,et al.  Mixed 2D/3D Perception for Autonomous Robots in Unstructured Environments , 2010, RoboCup.

[21]  A. N. Chernodub.  Training Neural Networks for classification using the Extended Kalman Filter: A comparative study , 2014, Optical Memory and Neural Networks.

[22]  Mark E. Oxley,et al.  Comparative Analysis of Backpropagation and the Extended Kalman Filter for Training Multilayer Perceptrons , 1992, IEEE Trans. Pattern Anal. Mach. Intell..

[23]  E. Baumgartner.  In-situ exploration of Mars using rover systems , 2000 .

[24]  Kazuya Yoshida,et al.  Path Planning and Evaluation for Planetary Rovers Based on Dynamic Mobility Index , 2011, 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[25]  Aaron C. Courville,et al.  Interacting Markov Random Fields for Simultaneous Terrain Modeling and Obstacle Detection , 2005, Robotics: Science and Systems.

[26]  Martial Hebert,et al.  Natural terrain classification using three‐dimensional ladar data for ground robot mobility , 2006, J. Field Robotics.

[27]  Timothy Masters,et al.  Practical neural network recipes in C , 1993 .

[28]  Nikola Kasabov,et al.  Foundations Of Neural Networks, Fuzzy Systems, And Knowledge Engineering [Books in Brief] , 1996, IEEE Transactions on Neural Networks.

[29]  Karl Iagnemma,et al.  Self‐supervised terrain classification for planetary surface exploration rovers , 2012, J. Field Robotics.

[30]  C. Sammut,et al.  Extracting Terrain Features from Range Images for Autonomous Random Stepfield Traversal , 2007, 2007 IEEE International Workshop on Safety, Security and Rescue Robotics.

[31]  Jorge L. Martinez,et al.  Fuzzy modeling of natural terrain elevation from a 3D scanner point cloud , 2011, 2011 IEEE 7th International Symposium on Intelligent Signal Processing.

[32]  Alex Ellery,et al.  Estimating terrain parameters for a rigid wheeled rover using neural networks , 2013 .

[33]  Ioannis M. Rekleitis,et al.  Autonomous planetary exploration using LIDAR data , 2009, 2009 IEEE International Conference on Robotics and Automation.

[34]  Timothy D. Barfoot,et al.  State Estimation for Robotics , 2017 .

[35]  Stephen I. Gallant,et al.  Neural network learning and expert systems , 1993 .

[36]  Tian Xia,et al.  Vehicle Detection from 3D Lidar Using Fully Convolutional Network , 2016, Robotics: Science and Systems.

[37]  Jürgen Schmidhuber,et al.  Deep learning in neural networks: An overview , 2014, Neural Networks.

[38]  Thomas B. Schön,et al.  Learning to close the loop from 3D point clouds , 2010, 2010 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[39]  Wolfram Burgard,et al.  Improving robot navigation in structured outdoor environments by identifying vegetation from laser data , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[40]  Anthony Stentz,et al.  Field D*: An Interpolation-Based Path Planner and Replanner , 2005, ISRR.

[41]  Joshua A. Marshall,et al.  Towards intensity-augmented SLAM with LiDAR and ToF sensors , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[42]  Larry H. Matthies,et al.  Vision Guided Landing of an Autonomous Helicopter in Hazardous Terrain , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[43]  Laurene V. Fausett,et al.  Fundamentals Of Neural Networks , 1993 .

[44]  Michael Beetz,et al.  Extending Player/Stage/Gazebo towards Cognitive Robots Acting in Ubiquitous Sensor-equipped Environments , 2007 .