| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
Within the lfit_bounding_box functions we take the unordered clusters, run a 2D PCA to then sort them with partial_sort() in order to be able to implement the algorithm seen in the paper "Efficient L-shape Fitting of Laser Scanner Data for Vehicle Pose Estimation". However, what I do not understand is how 2D PCA is being utilized to sort the point cloud. Can anyone explain this to me?
I believe I was able to find the answers I was looking for, instead of sorting the points radially as is hinted at in the paper, since the point cloud is not filtered Autoware made the decision to sort based of projection on the eigen value from 2D PCA.
Organizing it this way, when performing the index shift it should provide two perpendicular lines which least square fit the clustered set.
Asked: 2020-11-23 13:44:11 -0500
Seen: 227 times
Last updated: Nov 24 '20
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
Is this an autoware question?
If so: please label them as such.