ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange
Ask Your Question

Revision history [back]

Map making composes of two main modules .

  1. Actual mapiing
  2. Localization

Mapping is straight forward if you have a Kinect like sensor (RGB-D) , publish the topics and feed it to gmapping module . The tricky part is that with the Odometry , odometry can be achieved vai a couple of ways

  • Encoders + IMU fusion
  • Visual Odometry

Map making composes of two main modules .

  1. Actual mapiing
  2. Localization

Mapping is straight forward if you have a Kinect like sensor (RGB-D) , publish the topics and feed it to gmapping module . The tricky part is that with the Odometry , odometry can be achieved vai a couple of ways

  • Encoders + IMU fusion (EKF based)
  • Visual Odometry (rtabmap , ORB-SLAM2...)

If you are lucky enough to have a 360degree lidar , you can use hector mapping (This is a slam algorithm which works without encoder feedback )