ROS package for a simple navigation system utilizing the Visibility Graph algorithm.
It is compatible with occupancy grid maps generated by popular 2D SLAM such as Cartographer, slam_toolbox, and slam_gmapping

$ cd {ROS_WORKSPACE}/src $ git clone -b noetic-devel https://github.com/Alpaca-zip/vgraph_nav.git $ wstool merge vgraph_nav/.rosinstall $ wstool update $ python3 -m pip install -r vgraph_nav/requirements.txt $ rosdep install -r -y -i --from-paths . $ catkin build

Example of navigation using TurtleBot3.

$ roslaunch vgraph_nav turtlebot3_vgraph_navigation.launch 

Before starting navigation, you can define the starting point of the robot.
This is achieved using the 2D Pose Estimate button on the RViz.

More info : e-Manual

After setting the initial pose, the next step is to specify the target position using 2D Nav Goal.

More info : e-Manual

NOTE: The first time after launch, optimization using the Visibility Graph algorithm takes few minutes.

• ~map_file: Path to the configuration file (yaml) for the occupancy grid map.
• ~test_folder: A debug folder primarily used for debugging the Visibility Graph algorithm.
• ~epsilon_factor: Adjust the corner detection accuracy of the occupancy grid map. Higher values reduce the number of corners.
• ~robot_radius: Specifies the radius of the robot. This is mainly used for obstacle detection.
• ~odom_topic: The name of the odometry topic.
• ~scan_topic: The name of the LaserScan topic.
• ~vel_linear: Linear velocity.
• ~vel_theta: Maximum angular velocity.
• ~angle_tolerance: During navigation, the robot will attempt to align its direction with the goal if the angle difference exceeds this tolerance.
• ~goal_tolerance: Distance to a waypoint at which it is considered reached.
• ~verbose: If True, the results of the optimization are displayed.

• Subscribed Topics:

  • /initialpose (geometry_msgs/PoseWithCovarianceStamped) : Receives the robot's initial pose from RViz's 2D Pose Estimate
  • /move_base_simple/goal (geometry_msgs/PoseStamped) : Receives navigation goals from RViz's 2D Nav Goal
  • /odom (nav_msgs/Odometry) : Provides the robot's odometry data. Can be remapped via ~odom_topic.
  • /scan (sensor_msgs/LaserScan) : Provides the 2D Lidar data of the robot. Can be remapped via ~scan_topic.

• Published Topics:

  • /cost_map (nav_msgs/OccupancyGrid) : Generates a path along the obstacles in this occupancy grid map. Mainly for visualization in RViz.
  • /path (nav_msgs/Path) : The path that the robot will follow. Mainly for visualization in RViz.
  • /cmd_vel (geometry_msgs/Twist) : Conveys the target velocity data of the robot.