ground surface estimation

The LiDAR data and map data are sourced from the project: hdl_localization

Algorithm reference for this excellent project: pointcloud_surface

Introduction to Code Structure

Structure

-- cmake
​	# find third party as ${ThirdParty}.
-- config
	# save config yaml for system.
-- data
	# save examples point cloud map.
-- src
	# ground estimate
​	-- algorithm
		# surface estimate algorithm.
​	-- common
		# universal tools, struct, types for all project.
-- test
	# Specific usage for ground surface estimate.

Requirements

ground surface estimation requires the following libraries:

• PCL
• Eigen3
• glog gflags
• yaml-cpp
• ceres

Installation

git clone https://github.com/ZYCheng1002/ground_surface_estimation.git
./build_thirdparty.sh
mkdir build && cd build
cmake ..
make -j8

Evaluation

• Original point cloud visualization

• Ground fitting using SAC algorithm

• Ground fitting using polynominal(ceres)

Task List

• Overall universal structure
• [ ]
• Adapt Docker