An implementation of Gaussian Process Incremental Sample Consensus (GP-INSAC) algorithm
Author: Huang Pengdi
Email:alualu628628@163.com
This is an implementation of GP-INSAC algorithm in ROS system
The full name of GP-INSAC algorithm is Gaussian Process Incremental Sample Consensus algorithm
It is firstly proposed by B. Douillard et al. in the paper:
Douillard B, Underwood J, Kuntz N, et al, On the segmentation of 3D LIDAR point clouds, IEEE ICRA 2011, 2798-2805.
This algorithm is to obtain the travelable region (ground points) from a read-time point clouds (one frame online point cloud)
Note that the input topic of GP-INASC is macthing with the output topic of LOAM
Makesure you have a ubuntu 14.04 system and a ROS system ROS - Indigo, I think indigo or higher version is fine.
cd $YOUR_WORKSPACE_DIR$, and type:
catkin_make -DCMAKE_BUILD_TYPE=ReleaseMost of the algorithm parameters are declared in GP-INASC/launch/gpinsac.launch, including input topic and output topic.
Run this node after modifying $param name="lidar_topic"$ to your real-time point cloud topic:
roslaunch travelable_region gp_insac.launch If you still do not know how to deal with your configuration, here a example Husky_Simulation may be helpful, in which this node is used for real-time ground extraction.