Position Estimation on Image-Based Heat Map Input using Particle Filters in Cartesian Space

This paper presents an approach for using an image-based heat map as measurement input of a particle filter. Pixels of the heat map are transformed into Cartesian space relative to the robot and regarded as single measurements. The approach uses a novel observation model to weight the particles accordingly to the heat map pixels. While this paper focuses on handling FCNN output, the method is also applicable to other feature recognition methods like saliency approaches. The proposed method shows similar performance in standard cases and huge improvements on erroneous input compared to the conventional approach.

[1]  Raoul Daniel Zöllner,et al.  Image-based multi-target tracking using a multi-layer particle filter and extended EM clustering , 2017, 2017 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI).

[2]  Simone Frintrop,et al.  Salient Pattern Detection Using W 2 on Multivariate Normal Distributions , 2012, DAGM/OAGM Symposium.

[3]  Nanning Zheng,et al.  Automatic salient object segmentation based on context and shape prior , 2011, BMVC.

[4]  Norman Hendrich,et al.  ImageTagger: An Open Source Online Platform for Collaborative Image Labeling , 2018, RoboCup.

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

[6]  Davide Scaramuzza,et al.  Robot localization using soft object detection , 2012, 2012 IEEE International Conference on Robotics and Automation.

[7]  A. Hero,et al.  Multitarget tracking using the joint multitarget probability density , 2005, IEEE Transactions on Aerospace and Electronic Systems.

[8]  T. Başar,et al.  A New Approach to Linear Filtering and Prediction Problems , 2001 .

[9]  Tobi Delbrück,et al.  Combined frame- and event-based detection and tracking , 2016, 2016 IEEE International Symposium on Circuits and Systems (ISCAS).

[10]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[11]  Edwin Olson,et al.  AprilTag 2: Efficient and robust fiducial detection , 2016, 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).

[12]  Mark R. Morelande,et al.  A Bayesian Approach to Multiple Target Detection and Tracking , 2007, IEEE Transactions on Signal Processing.

[13]  K. Dorer,et al.  Detection and Localization of Features on a Soccer Field with Feedforward Fully Convolutional Neural Networks ( FCNN ) for the Adult-Size Humanoid Robot Sweaty , 2017 .

[15]  Michael S. Bernstein,et al.  ImageNet Large Scale Visual Recognition Challenge , 2014, International Journal of Computer Vision.

[16]  F. Ade,et al.  Using the condensation algorithm to implement tracking for mobile robots , 1999, 1999 Third European Workshop on Advanced Mobile Robots (Eurobot'99). Proceedings (Cat. No.99EX355).

[17]  N. Gordon,et al.  Novel approach to nonlinear/non-Gaussian Bayesian state estimation , 1993 .

[18]  Andrew W. Moore,et al.  X-means: Extending K-means with Efficient Estimation of the Number of Clusters , 2000, ICML.

[19]  Sergey Levine,et al.  Backprop KF: Learning Discriminative Deterministic State Estimators , 2016, NIPS.

[20]  Pablo V. A. Barros,et al.  Towards Real-Time Ball Localization Using CNNs , 2018, RoboCup.

[21]  Esther Koller-Meier,et al.  Tracking multiple objects using the Condensation algorithm , 2001, Robotics Auton. Syst..

[22]  Raoul Daniel Zöllner,et al.  An adaptive multi-layer particle filter for tracking of traffic participants in a roundabout , 2016, 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC).

[23]  Danylo Malyuta,et al.  Guidance, Navigation, Control and Mission Logic for Quadrotor Full-cycle Autonomy , 2018 .

[24]  J. MacQueen.  Some methods for classification and analysis of multivariate observations , 1967 .

[25]  Thomas Brox,et al.  U-Net: Convolutional Networks for Biomedical Image Segmentation , 2015, MICCAI.