Adaptive compression for 3 D laser data

This paper concerns the creation of efficient surface representations from laser point clouds created by a push broom laser system. We produce a continuous, implicit, non-parametric and non-stationary representation with an update time that is constant. This allows us to form predictions of the underlying workspace surfaces at arbitrary locations and densities. The algorithm places no restriction on the complexity of the surfaces and automatically prunes redundant data via an information theoretic criterion. This criterion makes the use of Gaussian Process Regression a natural choice. We adopt a formulation which handles the typical non-functional relation between XY location and elevation allowing us to map arbitrary environments. Results are presented that use real and synthetic data to analyse the trade-off between compression rate and reconstruction error. We attain decimation factors in excess of two orders of magnitude without significant degradation in fidelity.

[1]  Wolfram Burgard,et al.  An Efficient Extension of Elevation Maps for Outdoor Terrain Mapping , 2005, FSR.

[2]  Takeo Kanade,et al.  Ambler: an autonomous rover for planetary exploration , 1989, Computer.

[3]  A. O'Hagan,et al.  Bayes–Hermite quadrature , 1991 .

[4]  Wolfram Burgard,et al.  Gaussian Beam Processes: A Nonparametric Bayesian Measurement Model for Range Finders , 2007, Robotics: Science and Systems.

[5]  Neil D. Lawrence,et al.  Fast Forward Selection to Speed Up Sparse Gaussian Process Regression , 2003, AISTATS.

[6]  P. Newman,et al.  Generating Implicit Surfaces from Lidar Data , 2010 .

[7]  Marc Levoy,et al.  Efficient variants of the ICP algorithm , 2001, Proceedings Third International Conference on 3-D Digital Imaging and Modeling.

[8]  Wolfram Burgard,et al.  Adaptive Non-Stationary Kernel Regression for Terrain Modeling , 2007, Robotics: Science and Systems.

[9]  Wolfram Burgard,et al.  Learning predictive terrain models for legged robot locomotion , 2008, 2008 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[10]  Christian Früh,et al.  Constructing 3D City Models by Merging Aerial and Ground Views , 2003, IEEE Computer Graphics and Applications.

[11]  Wolfram Burgard,et al.  Multi-Level Surface Maps for Outdoor Terrain Mapping and Loop Closing , 2006, 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[12]  Jack Dongarra,et al.  LINPACK Users' Guide , 1987 .

[13]  P. Newman,et al.  Navigating , Recognising and Describing Urban Spaces With Vision and Laser , 2009 .

[14]  Wolfram Burgard,et al.  Learning compact 3D models of indoor and outdoor environments with a mobile robot , 2003, Robotics Auton. Syst..

[15]  Hans P. Moravec,et al.  Robot Evidence Grids. , 1996 .

[16]  Lehel Csató,et al.  Sparse On-Line Gaussian Processes , 2002, Neural Computation.

[17]  Carl E. Rasmussen,et al.  A Unifying View of Sparse Approximate Gaussian Process Regression , 2005, J. Mach. Learn. Res..

[18]  Michael A. Osborne.  Bayesian Gaussian processes for sequential prediction, optimisation and quadrature , 2010 .

[19]  Hugh F. Durrant-Whyte,et al.  Gaussian Process modeling of large scale terrain , 2009, 2009 IEEE International Conference on Robotics and Automation.

[20]  Winston Churchill,et al.  The New College Vision and Laser Data Set , 2009, Int. J. Robotics Res..

[21]  Carl E. Rasmussen,et al.  Gaussian processes for machine learning , 2005, Adaptive computation and machine learning.

[22]  Martial Hebert,et al.  Space-carving Kernels for Accurate Rough Terrain Estimation , 2010, Int. J. Robotics Res..

[23]  Tony DeRose,et al.  Mesh optimization , 1993, SIGGRAPH.

[24]  M. Seeger.  Low Rank Updates for the Cholesky Decomposition , 2004 .

[25]  Carl E. Rasmussen,et al.  Bayesian Monte Carlo , 2002, NIPS.

[26]  Sarvapali D. Ramchurn,et al.  2008 International Conference on Information Processing in Sensor Networks Towards Real-Time Information Processing of Sensor Network Data using Computationally Efficient Multi-output Gaussian Processes , 2022 .

[27]  G. Matheron.  The intrinsic random functions and their applications , 1973, Advances in Applied Probability.

[28]  Carl E. Rasmussen,et al.  Gaussian process dynamic programming , 2009, Neurocomputing.

[29]  R. Kohn,et al.  Estimation, Filtering, and Smoothing in State Space Models with Incompletely Specified Initial Conditions , 1985 .

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

[31]  Anthony O'Hagan,et al.  Monte Carlo is fundamentally unsound , 1987 .

[32]  Andreas Krause,et al.  Near-Optimal Sensor Placements in Gaussian Processes: Theory, Efficient Algorithms and Empirical Studies , 2008, J. Mach. Learn. Res..

[33]  Jan Peters,et al.  Using model knowledge for learning inverse dynamics , 2010, 2010 IEEE International Conference on Robotics and Automation.

[34]  Paul Newman,et al.  Efficient Non-Parametric Surface Representations Using Active Sampling for Push Broom Laser Data , 2010, Robotics: Science and Systems.

[35]  Wolfram Burgard,et al.  Using Hierarchical EM to Extract Planes from 3D Range Scans , 2005, Proceedings of the 2005 IEEE International Conference on Robotics and Automation.

[36]  David J. C. MacKay,et al.  Information Theory, Inference, and Learning Algorithms , 2004, IEEE Transactions on Information Theory.

[37]  A. O'Hagan,et al.  Curve Fitting and Optimal Design for Prediction , 1978 .