AVP-SLAM-PLUS is an implementation of AVP_SLAM and some new contributions. Performance of AVP-SLAM-PLUS could be found in video(https://www.bilibili.com/video/BV11R4y137xb/)
AVP-SLAM-PLUS contain a simple implementation of AVP-SLAM: Semantic Visual Mapping and Localization for Autonomous Vehicles in the Parking Lot(IROS 2020) and some new contributions.
The new contribustions are as follows: Firstly,the system provide two camera style mode which are multi RGB cameras mode and multi RGBD cameras mode; Secondly,the system provide two registration mode which are ICP mode and NDT mode. Lastly,the system provide mapping mode and localization mode, that means you can not only do SLAM,but also do localization in a prior map.
This code is simple and is a good learning material for SLAM beginners.
Author: Liu Guitao
Email: [email protected]
Ubuntu 64-bit 18.04.ROS Melodic. ROS Installation
cd ~/catkin_ws/src
git clone https://github.com/liuguitao/AVP-SLAM-PLUS.git
cd AVP-SLAM-PLUS/avp_slam_plus/model/
unzip my_ground_plane.zip -d ~/.gazebo/models/
cd ~/catkin_ws
catkin_make
source ~/catkin_ws/devel/setup.bash
if you want to save map and use the map to do localization, you should ensure your config file have be correctely set. The config file is at AVP-SLAM-PLUS/avp_slam_plus/configFile.yaml
mapSave: true
mapSaveLocation: your map file address
roslaunch avp_slam_plus slamRGB.launch
open a new terminal, control robot move
roslaunch robot_control robot_control.launch
if you have do 3.1.1 "save map", you can do localization in the prior map.
roslaunch avp_slam_plus localizaitonRGB.launch
open a new terminal, control robot move
roslaunch robot_control robot_control.launch
if you want to save map and use the map to do localization, you should ensure your config file have be correctely set. The config file is at AVP-SLAM-PLUS/avp_slam_plus/configFile.yaml
roslaunch avp_slam_plus slamRGBD.launch
open a new terminal, control robot move
roslaunch robot_control robot_control.launch
roslaunch avp_slam_plus localizaitonRGBD.launch
open a new terminal, control robot move
roslaunch robot_control robot_control.launch
Thanks for AVP_SLAM(Tong Qin, Tongqing Chen, Yilun Chen, and Qing Su:Semantic Visual Mapping and Localization for Autonomous Vehicles in the Parking Lot),
Thanks for TurtleZhong(https://github.com/TurtleZhong/AVP-SLAM-SIM), whose simulated environment help me a lot.
Thanks for huchunxu(https://github.com/huchunxu/ros_exploring), whose simulated robot model help me a lot.