Learning-Based Methods of Perception and Navigation for Ground Vehicles in Unstructured Environments: A Review

The problem of autonomous navigation of a ground vehicle in unstructured environments is both challenging and crucial for the deployment of this type of vehicle in real-world applications. Several well-established communities in robotics research deal with these scenarios such as search and rescue robotics, planetary exploration, and agricultural robotics. Perception plays a crucial role in this context, since it provides the necessary information to make the vehicle aware of its own status and its surrounding environment. We present a review on the recent contributions in the robotics literature adopting learning-based methods to solve the problem of environment perception and interpretation with the final aim of the autonomous context-aware navigation of ground vehicles in unstructured environments. To the best of our knowledge, this is the first work providing such a review in this context.

[1]  Alessandro Giusti,et al.  Learning Ground Traversability From Simulations , 2017, IEEE Robotics and Automation Letters.

[2]  Yuichi Kobayashi,et al.  Regressed Terrain Traversability Cost for Autonomous Navigation Based on Image Textures , 2020, Applied Sciences.

[3]  Xin Zhang,et al.  End to End Learning for Self-Driving Cars , 2016, ArXiv.

[4]  Javier Civera,et al.  The Rosario dataset: Multisensor data for localization and mapping in agricultural environments , 2018, Int. J. Robotics Res..

[5]  Juan Andrade-Cetto,et al.  Terrain Classification in Complex Three‐dimensional Outdoor Environments , 2015, J. Field Robotics.

[6]  Sebastian Scherer,et al.  Real-Time Semantic Mapping for Autonomous Off-Road Navigation , 2017, FSR.

[7]  Sanjiv Singh,et al.  The 2005 DARPA Grand Challenge: The Great Robot Race , 2007 .

[8]  Sonia Chernova,et al.  Recent Advances in Robot Learning from Demonstration , 2020, Annu. Rev. Control. Robotics Auton. Syst..

[9]  Ingmar Kanitscheider,et al.  Training recurrent networks to generate hypotheses about how the brain solves hard navigation problems , 2016, NIPS.

[10]  Klaus Obermayer,et al.  Support vector learning for ordinal regression , 1999 .

[11]  Alberto Broggi,et al.  Traversability analysis using terrain mapping and online-trained Terrain type classifier , 2014, 2014 IEEE Intelligent Vehicles Symposium Proceedings.

[12]  Richard Bowden,et al.  A Survey of Deep Learning Applications to Autonomous Vehicle Control , 2019, IEEE Transactions on Intelligent Transportation Systems.

[13]  Anh Nguyen,et al.  Autonomous Navigation in Complex Environments with Deep Multimodal Fusion Network , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[14]  Jan Faigl,et al.  On Unsupervised Learning of Traversal Cost and Terrain Types Identification Using Self-organizing Maps , 2019, ICANN.

[15]  Gaurav S. Sukhatme,et al.  Rover-IRL: Inverse Reinforcement Learning With Soft Value Iteration Networks for Planetary Rover Path Planning , 2018, IEEE Robotics and Automation Letters.

[16]  Byron Boots,et al.  Imitation learning for agile autonomous driving , 2019, Int. J. Robotics Res..

[17]  C. Qi.  Deep Learning on Point Sets for 3 D Classification and Segmentation , 2016 .

[18]  Suguru Arimoto,et al.  Bettering operation of Robots by learning , 1984, J. Field Robotics.

[19]  Huijing Zhao,et al.  Off-road Autonomous Vehicles Traversability Analysis and Trajectory Planning Based on Deep Inverse Reinforcement Learning , 2019, 2020 IEEE Intelligent Vehicles Symposium (IV).

[20]  Panagiotis Papadakis,et al.  Terrain traversability analysis methods for unmanned ground vehicles: A survey , 2013, Eng. Appl. Artif. Intell..

[21]  Brice Renaudeau,et al.  Air-ground cooperative topometric mapping of traversable ground , 2020, Auton. Robots.

[22]  Karl Iagnemma,et al.  DeepTerramechanics: Terrain Classification and Slip Estimation for Ground Robots via Deep Learning , 2018, ArXiv.

[23]  Nan Li,et al.  Predicting Terrain Mechanical Properties in Sight for Planetary Rovers with Semantic Clues , 2020, ArXiv.

[24]  Goldie Nejat,et al.  Robot Navigation of Environments with Unknown Rough Terrain Using deep Reinforcement Learning , 2018, 2018 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR).

[25]  Marco Pertile,et al.  Evaluation of 3D CNN Semantic Mapping for Rover Navigation , 2020, 2020 IEEE 7th International Workshop on Metrology for AeroSpace (MetroAeroSpace).

[26]  Matthew Zucker.  A Data-Driven Approach to High Level Planning , 2009 .

[27]  Pieter Abbeel,et al.  BADGR: An Autonomous Self-Supervised Learning-Based Navigation System , 2020, ArXiv.

[28]  Pål Johan From,et al.  End-to-end Learning for Autonomous Crop Row-following , 2019, IFAC-PapersOnLine.

[29]  Pieter Abbeel,et al.  Hierarchical Apprenticeship Learning with Application to Quadruped Locomotion , 2007, NIPS.

[30]  Marc Pollefeys,et al.  Semantic3D.net: A new Large-scale Point Cloud Classification Benchmark , 2017, ArXiv.

[31]  Alberto Ferreira de Souza,et al.  Self-Driving Cars: A Survey , 2019, Expert Syst. Appl..

[32]  Jari Saarinen,et al.  Normal Distributions Transform Traversability Maps: LIDAR‐Only Approach for Traversability Mapping in Outdoor Environments , 2017, J. Field Robotics.

[33]  Myungjin Jung,et al.  Vision-Based Terrain Classification and Solar Irradiance Mapping for Solar-Powered Robotics , 2018, 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[34]  Markus Wulfmeier,et al.  Maximum Entropy Deep Inverse Reinforcement Learning , 2015, 1507.04888.

[35]  Nicholas Roy,et al.  Nonparametric Bayesian inference on multivariate exponential families , 2014, NIPS.

[36]  Vignesh Sushrutha Raghavan,et al.  Terrain Segmentation and Roughness Estimation using RGB Data: Path Planning Application on the CENTAURO Robot , 2019, 2019 IEEE-RAS 19th International Conference on Humanoid Robots (Humanoids).

[37]  Razvan Pascanu,et al.  Learning to Navigate in Complex Environments , 2016, ICLR.

[38]  Lorenz Wellhausen,et al.  Learning quadrupedal locomotion over challenging terrain , 2020, Science Robotics.

[39]  Roland Siegwart,et al.  From perception to decision: A data-driven approach to end-to-end motion planning for autonomous ground robots , 2016, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[40]  David J. Brady,et al.  Multiscale gigapixel photography , 2012, Nature.

[41]  Leonidas J. Guibas,et al.  PointNet: Deep Learning on Point Sets for 3D Classification and Segmentation , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[42]  Dushyant Rao,et al.  Large-scale cost function learning for path planning using deep inverse reinforcement learning , 2017, Int. J. Robotics Res..

[43]  Giulio Reina,et al.  Learning Traversability From Point Clouds in Challenging Scenarios , 2018, IEEE Transactions on Intelligent Transportation Systems.

[44]  Jorge L. Martinez,et al.  Supervised Learning of Natural-Terrain Traversability with Synthetic 3D Laser Scans , 2020 .

[45]  David Silver,et al.  Learning to search: Functional gradient techniques for imitation learning , 2009, Auton. Robots.

[46]  Filip Marić,et al.  The Canadian Planetary Emulation Terrain Energy-Aware Rover Navigation Dataset , 2020, Int. J. Robotics Res..

[47]  Shirel Josef,et al.  Deep Reinforcement Learning for Safe Local Planning of a Ground Vehicle in Unknown Rough Terrain , 2020, IEEE Robotics and Automation Letters.

[48]  Sebastian Scherer,et al.  Integrating kinematics and environment context into deep inverse reinforcement learning for predicting off-road vehicle trajectories , 2018, CoRL.

[49]  T. Delbruck,et al.  A 128 128 120 dB 15 s Latency Asynchronous Temporal Contrast Vision Sensor , 2006 .

[50]  Giovanni Muscato,et al.  Domain Adaptation for Outdoor Robot Traversability Estimation from RGB data with Safety-Preserving Loss , 2020, 2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[51]  Xi Chen,et al.  Geometric and visual terrain classification for autonomous mobile navigation , 2017, 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[52]  Roland Siegwart,et al.  Navigation planning for legged robots in challenging terrain , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[53]  Wolfram Burgard,et al.  Towards Robust Semantic Segmentation using Deep Fusion , 2010 .

[54]  Kiho Kwak,et al.  Probabilistic traversability map generation using 3D-LIDAR and camera , 2016, 2016 IEEE International Conference on Robotics and Automation (ICRA).

[55]  Leonidas J. Guibas,et al.  PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Space , 2017, NIPS.

[56]  Krzysztof Walas,et al.  Where Should I Walk? Predicting Terrain Properties From Images Via Self-Supervised Learning , 2019, IEEE Robotics and Automation Letters.

[57]  M. Botvinick,et al.  The hippocampus as a predictive map , 2016 .

[58]  Andrew Y. Ng,et al.  A control architecture for quadruped locomotion over rough terrain , 2008, 2008 IEEE International Conference on Robotics and Automation.

[59]  Robert A. Hewitt,et al.  Training a terrain traversability classifier for a planetary rover through simulation , 2017 .

[60]  Toby P. Breckon,et al.  Learning to Drive: Using Visual Odometry to Bootstrap Deep Learning for Off-Road Path Prediction , 2018, 2018 IEEE Intelligent Vehicles Symposium (IV).

[61]  Weidong Cao,et al.  A Survey on Theories and Applications for Self-Driving Cars Based on Deep Learning Methods , 2020, Applied Sciences.

[62]  Stefano Ermon,et al.  InfoGAIL: Interpretable Imitation Learning from Visual Demonstrations , 2017, NIPS.

[63]  Angela P. Schoellig,et al.  Visual teach and repeat, repeat, repeat: Iterative Learning Control to improve mobile robot path tracking in challenging outdoor environments , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[64]  Christopher G. Atkeson,et al.  Optimization and learning for rough terrain legged locomotion , 2011, Int. J. Robotics Res..

[65]  Owen Holland.  The first biologically inspired robots , 2003, Robotica.

[66]  Chao He,et al.  Vision-Based Real-Time Traversable Region Detection for Mobile Robot in the Outdoors , 2017, Sensors.

[67]  Martin A. Riedmiller,et al.  Towards General and Autonomous Learning of Core Skills: A Case Study in Locomotion , 2020, CoRL.

[68]  Paolo Valigi,et al.  Deep Reinforcement Learning for Instruction Following Visual Navigation in 3D Maze-Like Environments , 2020, IEEE Robotics and Automation Letters.

[69]  Wolfram Burgard,et al.  The limits and potentials of deep learning for robotics , 2018, Int. J. Robotics Res..

[70]  Dushyant Rao,et al.  Vote3Deep: Fast object detection in 3D point clouds using efficient convolutional neural networks , 2016, 2017 IEEE International Conference on Robotics and Automation (ICRA).

[71]  Gregory Dudek,et al.  Learning to Drive Off Road on Smooth Terrain in Unstructured Environments Using an On-Board Camera and Sparse Aerial Images , 2020, 2020 IEEE International Conference on Robotics and Automation (ICRA).

[72]  Wolfram Burgard,et al.  Traversability analysis for mobile robots in outdoor environments: A semi-supervised learning approach based on 3D-lidar data , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[73]  David M. Bradley,et al.  Boosting Structured Prediction for Imitation Learning , 2006, NIPS.

[74]  Sorin Grigorescu,et al.  A Survey of Deep Learning Techniques for Autonomous Driving , 2020, J. Field Robotics.

[75]  Daniel Matolin,et al.  An asynchronous time-based image sensor , 2008, 2008 IEEE International Symposium on Circuits and Systems.

[76]  David Silver,et al.  Learning from Demonstration for Autonomous Navigation in Complex Unstructured Terrain , 2010, Int. J. Robotics Res..

[77]  Brian H. Wilcox,et al.  Sojourner on Mars and Lessons Learned for Future Plantery Rovers , 1998 .

[78]  Douglas Hackett,et al.  An overview of the Defense Advanced Research Projects Agency’s Learning Locomotion program , 2011, Int. J. Robotics Res..

[79]  Matthew P. Castanier,et al.  Off‐road ground robot path energy cost prediction through probabilistic spatial mapping , 2020, J. Field Robotics.

[80]  Kiho Kwak,et al.  An incremental nonparametric Bayesian clustering-based traversable region detection method , 2017, Auton. Robots.

[81]  Manuela M. Veloso,et al.  Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation , 2017, CoRL.

[82]  Rémi Gilleron,et al.  Text Classification from Positive and Unlabeled Examples , 2002 .

[83]  Wei Gao,et al.  Intention-Net: Integrating Planning and Deep Learning for Goal-Directed Autonomous Navigation , 2017, CoRL.

[84]  Oliver Brock,et al.  Mobility and Manipulation , 2016, Springer Handbook of Robotics, 2nd Ed..

[85]  Harry M. Jol,et al.  Ground penetrating radar : theory and applications , 2009 .

[86]  Sergey Levine,et al.  Self-Supervised Deep Reinforcement Learning with Generalized Computation Graphs for Robot Navigation , 2017, 2018 IEEE International Conference on Robotics and Automation (ICRA).

[87]  Wolfram Burgard,et al.  A Survey of Deep Network Solutions for Learning Control in Robotics: From Reinforcement to Imitation , 2016 .

[88]  Jan Peters,et al.  Reinforcement learning in robotics: A survey , 2013, Int. J. Robotics Res..

[89]  Quan Pan,et al.  A survey on multi-sensor fusion based obstacle detection for intelligent ground vehicles in off-road environments , 2020, Frontiers of Information Technology & Electronic Engineering.

[90]  Chengchao Bai,et al.  Three-Dimensional Vibration-Based Terrain Classification for Mobile Robots , 2019, IEEE Access.

[91]  Yann LeCun,et al.  Off-Road Obstacle Avoidance through End-to-End Learning , 2005, NIPS.

[92]  Christopher Brunskill,et al.  Terrain trafficability analysis and soil mechanical property identification for planetary rovers: A survey , 2012 .

[93]  Markus Wulfmeier,et al.  On Machine Learning and Structure for Mobile Robots , 2018, ArXiv.

[94]  Sergey Levine,et al.  Composable Action-Conditioned Predictors: Flexible Off-Policy Learning for Robot Navigation , 2018, CoRL.

[95]  Emilio Frazzoli,et al.  A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles , 2016, IEEE Transactions on Intelligent Vehicles.

[96]  Masahiro Ono,et al.  SPOC: Deep Learning-based Terrain Classification for Mars Rover Missions , 2016 .

[97]  Charles Elkan,et al.  Learning classifiers from only positive and unlabeled data , 2008, KDD.

[98]  Sergey Levine,et al.  Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimization , 2016, ICML.

[99]  Donald B. Gennery,et al.  Traversability Analysis and Path Planning for a Planetary Rover , 1999, Auton. Robots.

[100]  Anatoliy Boryssenko.  Preventing Damage by Hidden Objects in Vegetation , 2006, IEEE Aerospace and Electronic Systems Magazine.

[101]  Brendan Englot,et al.  Bayesian Generalized Kernel Inference for Terrain Traversability Mapping , 2018, CoRL.

[102]  Giovanni Muscato,et al.  Coverage path planning for a flock of aerial vehicles to support autonomous rovers through traversability analysis , 2019 .