LiDAR based relative pose and covariance estimation for communicating vehicles exchanging a polygonal model of their shape

Relative localization between autonomous vehicles is an important issue for accurate cooperative localization. It is also essential for obstacle avoidance or platooning. Thanks to communication between vehicles, additional information, such as vehicle model and dimension, can be transmitted to facilitate this relative localization process. In this paper, we present and compare different algorithms to solve this problem based on LiDAR points and the pose and model communicated by another vehicle. The core part of the algorithm relies on iterative minimization tested with two methods and different model associations using point-to-point and point-to-line distances. This work compares the accuracy, the consistency and the number of iterations needed to converge for the different algorithms in different scenarios, e.g. straight lane, two lanes and curved lane driving.

[1]  Sebastian Thrun,et al.  Model based vehicle detection and tracking for autonomous urban driving , 2009, Auton. Robots.

[2]  Weisi Lin,et al.  A closed-form estimate of 3D ICP covariance , 2015, 2015 14th IAPR International Conference on Machine Vision Applications (MVA).

[3]  Xiao Zhang,et al.  Efficient L-shape fitting for vehicle detection using laser scanners , 2017, 2017 IEEE Intelligent Vehicles Symposium (IV).

[4]  Kok-Lim Low Linear Least-Squares Optimization for Point-to-Plane ICP Surface Registration , 2004 .

[5]  Albert-Jan Baerveldt,et al.  Robot localization based on scan-matching - estimating the covariance matrix for the IDC algorithm , 2003, Robotics Auton. Syst..

[6]  Thierry Chateau,et al.  Deep MANTA: A Coarse-to-Fine Many-Task Network for Joint 2D and 3D Vehicle Analysis from Monocular Image , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Akshay Prabhakar Shetty,et al.  GPS-LiDAR sensor fusion aided by 3D city models for UAVs , 2017 .

[8]  Andrea Censi,et al.  An accurate closed-form estimate of ICP's covariance , 2007, Proceedings 2007 IEEE International Conference on Robotics and Automation.