Learning-Based Extended Object Tracking Using Hierarchical Truncation Measurement Model With Automotive Radar

This paper presents a data-driven measurement model for extended object tracking (EOT) with automotive radar. Specifically, the spatial distribution of automotive radar measurements is modeled as a hierarchical truncated Gaussian (HTG) with structural geometry parameters that can be learned from the training data. The HTG measurement model provides an adequate resemblance to the spatial distribution of real-world automotive radar measurements. Moreover, large-scale radar datasets can be leveraged to learn the geometry-related model parameters and offload the computationally demanding model parameter estimation from the state update step. The learned HTG measurement model is further incorporated into a random matrix based EOT approach with two (multi-sensor) measurement updates: one is based on a factorized Gaussian inverse-Wishart density representation and the other is based on a Rao-Blackwellized particle density representation. The effectiveness of the proposed approaches is verified on both synthetic data and real-world nuScenes dataset over 300 trajectories.

[1]  Klaus C. J. Dietmayer,et al.  A direct scattering model for tracking vehicles with high-resolution radars , 2016, 2016 IEEE Intelligent Vehicles Symposium (IV).

[2]  Karl Granström,et al.  Poisson Multi-Bernoulli Mixture Conjugate Prior for Multiple Extended Target Filtering , 2016, IEEE Transactions on Aerospace and Electronic Systems.

[3]  Jun S. Liu,et al.  Blind Deconvolution via Sequential Imputations , 1995 .

[4]  L. Hammarstrand,et al.  Extended Object Tracking using a Radar Resolution Model , 2012, IEEE Transactions on Aerospace and Electronic Systems.

[5]  Marcus Baum,et al.  Optimal Fusion of Elliptic Extended Target Estimates based on the Wasserstein Distance , 2019, 2019 22th International Conference on Information Fusion (FUSION).

[6]  C. Givens,et al.  A class of Wasserstein metrics for probability distributions. , 1984 .

[7]  Karl Granström,et al.  Metrics for performance evaluation of elliptic extended object tracking methods , 2016, 2016 IEEE International Conference on Multisensor Fusion and Integration for Intelligent Systems (MFI).

[8]  Qiang Xu,et al.  nuScenes: A Multimodal Dataset for Autonomous Driving , 2019, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[10]  Shunqiao Sun,et al.  MIMO Radar for Advanced Driver-Assistance Systems and Autonomous Driving: Advantages and Challenges , 2020, IEEE Signal Processing Magazine.

[11]  Gerd Wanielik,et al.  Probabilistic data association for tracking extended targets under clutter using random matrices , 2015, 2015 18th International Conference on Information Fusion (Fusion).

[12]  Jürgen Dickmann,et al.  The volcanormal density for radar-based extended target tracking , 2017, 2017 IEEE 20th International Conference on Intelligent Transportation Systems (ITSC).

[13]  LI X.RONG,et al.  Survey of maneuvering target tracking. Part I. Dynamic models , 2003 .

[14]  Ratnasingham Tharmarasa,et al.  Extended Target Tracking With Multipath Detections, Terrain-Constrained Motion Model and Clutter , 2020 .

[15]  Petros Boufounos,et al.  Extended Object Tracking with Automotive Radar Using Learned Structural Measurement Model , 2020, 2020 IEEE Radar Conference (RadarConf20).

[16]  Peter Willett,et al.  Consistent Linear Tracker With Converted Range, Bearing, and Range Rate Measurements , 2017, IEEE Transactions on Aerospace and Electronic Systems.

[17]  Adrian Wills,et al.  A Closed-Form Prediction Update for Extended Target Tracking Using Random Matrices , 2020, IEEE Transactions on Signal Processing.

[18]  Karl Granström,et al.  A phd Filter for Tracking Multiple Extended Targets Using Random Matrices , 2012, IEEE Transactions on Signal Processing.

[19]  Umut Orguner,et al.  A Variational Measurement Update for Extended Target Tracking With Random Matrices , 2012, IEEE Transactions on Signal Processing.

[20]  Marcus Baum,et al.  A Cartesian B-Spline Vehicle Model for Extended Object Tracking , 2018, 2018 21st International Conference on Information Fusion (FUSION).

[21]  Karl Granström,et al.  Extended Object Tracking: Introduction, Overview and Applications , 2016, ArXiv.

[22]  Paolo Braca,et al.  Multistatic Bayesian extended target tracking , 2016, IEEE Transactions on Aerospace and Electronic Systems.

[23]  Toshiaki Koike-Akino,et al.  Extended Object Tracking Using Hierarchical Truncation Measurement Model with Automotive Radar , 2020, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[24]  Uwe D. Hanebeck,et al.  Extended Object Tracking with Random Hypersurface Models , 2013, IEEE Transactions on Aerospace and Electronic Systems.

[25]  Marcus Baum,et al.  Tracking the Orientation and Axes Lengths of an Elliptical Extended Object , 2019, IEEE Transactions on Signal Processing.

[26]  Hans-Joachim Wünsche,et al.  Radar reflection characteristics of vehicles for contour and feature estimation , 2017, 2017 Sensor Data Fusion: Trends, Solutions, Applications (SDF).

[27]  Klaus C. J. Dietmayer,et al.  Tracking Multiple Vehicles Using a Variational Radar Model , 2017, IEEE Transactions on Intelligent Transportation Systems.

[28]  Jürgen Dickmann,et al.  Probabilistic rectangular-shape estimation for extended object tracking , 2016, 2016 IEEE Intelligent Vehicles Symposium (IV).

[29]  Philipp Berthold,et al.  An Abstracted Radar Measurement Model for Extended Object Tracking , 2018, 2018 21st International Conference on Intelligent Transportation Systems (ITSC).

[30]  Lennart Svensson,et al.  Adaptive Radar Sensor Model for Tracking Structured Extended Objects , 2012, IEEE Transactions on Aerospace and Electronic Systems.

[31]  Yu Liu,et al.  Extended Object Tracking Using Automotive Radar , 2018, 2018 21st International Conference on Information Fusion (FUSION).

[32]  Marcus Baum,et al.  Extended Target Tracking Using Gaussian Processes with High-Resolution Automotive Radar , 2018, 2018 21st International Conference on Information Fusion (FUSION).

[33]  Petros Boufounos,et al.  Extended Object Tracking Using Hierarchical Truncation Model with Partial-View Measurements , 2020 .

[34]  Paolo Braca,et al.  Multiple Sensor Measurement Updates for the Extended Target Tracking Random Matrix Model , 2017, IEEE Transactions on Aerospace and Electronic Systems.

[35]  Hauke Kaulbersch,et al.  EM-based Extended Target Tracking with Automotive Radar using Learned Spatial Distribution Models , 2019, 2019 22th International Conference on Information Fusion (FUSION).

[36]  J.W. Koch,et al.  Bayesian approach to extended object and cluster tracking using random matrices , 2008, IEEE Transactions on Aerospace and Electronic Systems.

[37]  Fredrik Gustafsson,et al.  Bayesian Inference via Approximation of Log-likelihood for Priors in Exponential Family , 2015, ArXiv.

[38]  Dietrich Fränken,et al.  Tracking of Extended Objects and Group Targets Using Random Matrices , 2008, IEEE Transactions on Signal Processing.