Skip to content
forked from leokoppel/msckf

Multi-State Constraint Kalman Filter with ROS interface.

Notifications You must be signed in to change notification settings

TBetterman/msckf-1

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

MSCKF

Experimental implementation of the Multi-State Constraint Kalman Filter with ROS interface.

Based on the paper by Mourikis and Roumeliotis (PDF).

Also includes a feature tracking node using OpenCV's SIFT implementation.

This is a work in progress. In particular, The API is a bit non-standard and is subject to change.

ROS nodes

feature_tracking_node

Subscribed topics

  • camera/image_rect (sensor_msgs/Image) - Undistorted image stream.
  • camera/camera_info (sensor_msgs/CameraInfo) - The corresponding camera metadata.

Published topics

  • output_image (sensor_msgs/Image) - Image with arrows drawn on it, showing matches.
  • features (msckf/ImageFeatures) - custom message describing feature positions

msckf_node

Subscribed topics

  • imu_vel (geometry_msgs/TwistStamped) - IMU measurement of linear and angular velocity.
  • odom (nav_msgs/Odometry) - Position of robot, used to initialize the filter
  • camera/image_rect (sensor_msgs/Image)- the same image stream used by feature_tracking_node. Currently, only the header information is used.
  • features (msckf/ImageFeatures) - the feature information published by feature_tracking_node

Published topics

  • odom_combined (geometry_msgs/PoseWithCovarianceStamped) - Estimated robot pose

About

Multi-State Constraint Kalman Filter with ROS interface.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • C++ 88.0%
  • Python 9.1%
  • CMake 2.9%