Only released in EOL distros:
- Author: Narcis palomeras
- License: BSD
- Source: git https://github.com/narcispr/pose_ekf_slam (branch: None)
Package pose-ekf-slam is a modular localization and mapping system for 6DoF vehicles. Standard navigation messages as PoseWithCovarianceStamped, TwistWithCovarianceStamped, and Imu sensor messages are combined by means of an extended Kalman filter. This node is capable of estimating vehicle position and linear velocity as well as the position of detected landmarks in real-time. It is entirely written in python for ROS Fuerte. It's philosophy is similar to package robot_pose_ekf but with several enchancements.
- Author: Narcís Palomeras
- License: BSD
Source: git https://github.com/narcispr/pose_ekf_slam
How to use Pose EKF-SLAM
Package pose_ekf_slam contains two nodes:
This node receive position fixes of type PoseWithCovarianceStamped in the topic
linear velocity updates of type TwistWithCovarianceStamped are received through the topic
and vehicle orientation and angular velocity are expected at
using an Imu sensor message. Finally, landmark updates have to be published at
in order to be discovered by the pose_ekf_slam node. They are also PoseWithCovarianceStamped messages.
Outputs of this node are an Odometry message published at
and a custom message, indicating the position and covariance of each landmark, published at
This node can be used to test pose_ekf_slam but the main idea is to be modified for every user in order to transform their navigation messages, with position, twist or landmarks information, into the kind of messages expected by the pose_ekf_slam node.
More information about this node can be found at:
N. Palomeras, S. Nagappa, D. Ribas, N. Gracias and M. Carreras. Vision-based localization and mapping system for AUV intervention. OCEANS'13 MTS/IEEE. Bergen, 2013.