mrsp / g2o_slam3d

g2o_slam3d: A ROS/C++ Visual SLAM with g2o

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.

Prerequisites

• Ubuntu 16.04 and later
• ROS kinetic and later
• Eigen 3.3.0 and later
• OpenCV 3.3 and later
• g2o

Installing

• git clone https://github.com/mrsp/g2o_slam3d.git
• catkin_make

Launch

• Specify configuration parameters in a yaml file e.g. in config/intel.yaml
• roslaunch g2o_slam g2o_slam_intel.launch
  (To launch the g2o with 3D Landmarks derived from 2D RGB features + Depth and an (optional) odometry measurements)
• roslaunch g2o_vslam g2o_vslam_intel.launch
  (To launch the g2o Bundle Adjustment with 2D RGB features and an (optional) odometry measurements)

Example of 3D VSLAM

Red is the Visual Odometry, Green is the Keyframes trajectory, frames indicate the estimated poses of Keyframes

Example of BA

Red is the Visual Odometry, Green is the Keyframes trajectory, frames indicate the estimated poses of Keyframes

Without ROS

BA with 2 RGB images

./baRGB ../am1.png ../am2.png

BA with 2 RGB and 2 Depth Images

./baRGBD ../am1.png ../am2.png ../amd1.png ../amd2.png