Map Summarization for Tractable Lifelong Mapping

Accurate pose estimation has become a core capability for modern robotics, required in a multitude of applications, where precise navigation or collaboration is involved. In beaconfree and GPS-denied environments, visual-based odometry and localization systems are a popular solution for obtaining reliable, high-frequency pose estimates. The current state-of-the-art approaches, however, are limited to rather short timespans and small scales due to the excessive amount of data which is being produced in each mapping session. This issue is particularly relevant if a wireless online communication between mapping peers is required, where bandwidth is limited. This paper investigates map summarization and reducing the data flow in visual feature-based localization systems. We examine the data storage and transmission requirements of stateof-the-art localization systems and identify potential remedies to existing limitations. We discuss choices and methods related to the localization map representation, lifelong map maintenance and landmark selection approaches. Finally, we evaluate some of the discussed methods in a real-life scenario, using an autonomous mobile robot repeatedly mapping an ever-changing office space over a period of 2 months. We prove it is possible to reduce the data transfer by a factor of 5 while maintaining good localization performance.

[1]  Michael Bosse,et al.  The gist of maps - summarizing experience for lifelong localization , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[2]  Michael Kaess,et al.  Generic Node Removal for Factor-Graph SLAM , 2014, IEEE Transactions on Robotics.

[3]  Torsten Sattler,et al.  Fast image-based localization using direct 2D-to-3D matching , 2011, 2011 International Conference on Computer Vision.

[4]  Laurent Kneip,et al.  Collaborative monocular SLAM with multiple Micro Aerial Vehicles , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[5]  Michael Bosse,et al.  Summary Maps for Lifelong Visual Localization , 2016, J. Field Robotics.

[6]  Yaser Sheikh,et al.  3D Point Cloud Reduction Using Mixed-Integer Quadratic Programming , 2013, 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops.

[7]  Luc Van Gool,et al.  Matching Features Correctly through Semantic Understanding , 2014, 2014 2nd International Conference on 3D Vision.

[8]  Juan Andrade-Cetto,et al.  Information-Based Compact Pose SLAM , 2010, IEEE Transactions on Robotics.

[9]  Horst Wildenauer,et al.  Descriptor free visual indoor localization with line segments , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[10]  Paul Newman,et al.  FAB-MAP: Probabilistic Localization and Mapping in the Space of Appearance , 2008, Int. J. Robotics Res..

[11]  Kurt Konolige,et al.  Towards lifelong visual maps , 2009, 2009 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[12]  Michael Milford,et al.  Place Recognition with ConvNet Landmarks: Viewpoint-Robust, Condition-Robust, Training-Free , 2015, Robotics: Science and Systems.

[13]  Winston Churchill,et al.  Experience-based navigation for long-term localisation , 2013, Int. J. Robotics Res..

[14]  Michael Bosse,et al.  Keypoint design and evaluation for place recognition in 2D lidar maps , 2009, Robotics Auton. Syst..

[15]  Jeffrey A. Hoffman,et al.  Graph-based terrain relative navigation with optimal landmark database selection , 2015, 2015 IEEE Aerospace Conference.

[16]  Edwin Olson,et al.  AprilTag: A robust and flexible visual fiducial system , 2011, 2011 IEEE International Conference on Robotics and Automation.

[17]  Jonathan P. How,et al.  Two-Stage Focused Inference for Resource-Constrained Collision-Free Navigation , 2015, Robotics: Science and Systems.

[18]  Stergios I. Roumeliotis,et al.  C-KLAM: Constrained keyframe-based localization and mapping , 2014, 2014 IEEE International Conference on Robotics and Automation (ICRA).

[19]  Konrad Schindler,et al.  Predicting Matchability , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[20]  Simon Lacroix,et al.  Probabilistic place recognition with covisibility maps , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[21]  Michael Bosse,et al.  Placeless Place-Recognition , 2014, 2014 2nd International Conference on 3D Vision.

[22]  Michael Bosse,et al.  Keep it brief: Scalable creation of compressed localization maps , 2015, 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[23]  H. Bischof,et al.  From structure-from-motion point clouds to fast location recognition , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[24]  Roland Siegwart,et al.  BRISK: Binary Robust invariant scalable keypoints , 2011, 2011 International Conference on Computer Vision.

[25]  John J. Leonard,et al.  Dynamic pose graph SLAM: Long-term mapping in low dynamic environments , 2012, 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[26]  Roland Siegwart,et al.  A synchronized visual-inertial sensor system with FPGA pre-processing for accurate real-time SLAM , 2014, ICRA 2014.

[27]  Torsten Sattler,et al.  Scalable 6-DOF Localization on Mobile Devices , 2014, ECCV.

[28]  Michael Bosse,et al.  Get Out of My Lab: Large-scale, Real-Time Visual-Inertial Localization , 2015, Robotics: Science and Systems.

[29]  Michael Bosse,et al.  Keyframe-based visual–inertial odometry using nonlinear optimization , 2015, Int. J. Robotics Res..

[30]  Cordelia Schmid,et al.  Aggregating local descriptors into a compact image representation , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[31]  John J. Leonard,et al.  Location utility-based map reduction , 2015, 2015 IEEE International Conference on Robotics and Automation (ICRA).

[32]  Luc Van Gool,et al.  Scale-invariant line descriptors for wide baseline matching , 2014, IEEE Winter Conference on Applications of Computer Vision.

[33]  Tom Drummond,et al.  Faster and Better: A Machine Learning Approach to Corner Detection , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[34]  Wolfram Burgard,et al.  Robust landmark selection for mobile robot navigation , 2013, 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems.

[35]  Dario Maio,et al.  Saliency-based keypoint selection for fast object detection and matching , 2015, Pattern Recognit. Lett..

[36]  Andrew Zisserman,et al.  Extremely Low Bit-Rate Nearest Neighbor Search Using a Set Compression Tree , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[37]  Matthijs C. Dorst.  Distinctive Image Features from Scale-Invariant Keypoints , 2011 .