The site is read-only. Please transition to use Robotics Stack Exchange
| ROS Resources: Documentation | Support | Discussion Forum | Index | Service Status | ros @ Robotics Stack Exchange |
 | 1 | initial version |
Hi @Shivendra,
The difference you see between Gazebo and the map is actually the normal behaviour when you launch the navigation for the first time. Since you are using AMCL to localize the robot in the map, you need to provide AMCL with a first pose estimate in order to begin the localization process.
This can be done by clicking in the top Panel of Rviz named 2D Pose Estimate and clicking again in the map in the "real" pose the robot is placed in the Gazebo world. You will see how AMCL corrects its estimations and the Pose array provided significantly decreases the uncertainty.
| | 2 | No.2 Revision |
Hi @Shivendra,
The difference you see between Gazebo and the map is actually the normal behaviour when you launch the navigation for the first time. Since you are using AMCL AMCL to localize the robot in the map, you need to provide AMCL with a first pose estimate in order to begin the localization process.
This can be done by clicking in the top Panel of Rviz named 2D Pose Estimate and clicking again in the map in the "real" pose the robot is placed in the Gazebo world. You will see how AMCL corrects its estimations and the Pose array provided significantly decreases the uncertainty.
ROS Answers is licensed under Creative Commons Attribution 3.0 Content on this site is licensed under a Creative Commons Attribution Share Alike 3.0 license.
