https://github.com/osll/slam-dataset-generation
Repository for a set of tools used to generate simulated datasets for use in developing SLAM algorithms
https://github.com/osll/slam-dataset-generation
Last synced: 14 days ago
JSON representation
Repository for a set of tools used to generate simulated datasets for use in developing SLAM algorithms
- Host: GitHub
- URL: https://github.com/osll/slam-dataset-generation
- Owner: OSLL
- Created: 2018-06-06T10:19:49.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2018-09-13T09:03:02.000Z (almost 8 years ago)
- Last Synced: 2025-02-21T14:15:39.304Z (over 1 year ago)
- Language: C++
- Size: 2.65 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Dataset generation tool for SLAM in ROS
This tools could help you to make a dataset for testing SLAM algorithms in ROS
How do I use this tool?
## Step 1: Create a map file
- Map files are svg files populated with [paths](https://developer.mozilla.org/en-US/docs/Web/SVG/Tutorial/Paths) that can be given ids
- Every map file must contain **at least two** special paths
* **world_boundary** = a closed path that represents the outermost boundary of an environment
+ Must be given "world_boundary" as an ID or it will not be recognized
* **robot_trajectory** = a path that represents the intermediate positions on the robot's trajectory through the environment
+ Must be given "robot_trajectory" as an ID or it will not be recognized
- In addition to those two paths, additional obstacles can be created
* Obstacles must be closed (*i.e.* the beginning and the end of the path must be the same)
- Any path other than the robot_trajectory can be given a trajectory that describes the motion of the obstacle
* To give an obstacle a trajectory
1) Associate the obstacle path with a unique id(inkscape typically does this by default)
2) Create the trajectory path and give it an id in the form of "${obstacle_id}_trajectory"
+ If a trajectory is detected but a corresponding obstacle to be associated with couldn't be found, then the parsing will fail with an error
## Step 2: Build the project
- Note that this project depends on boost
- Steps:
1) Navigate to the base of the repo
2) Issue `catkin_make`
## Step 3: Run the tool on the map file
- Steps:
1) Navigate to the base of the repo
2) Issue `source ./devel/setup.bash`
3) Issue `rosrun slam-dataset-generation simulate_world `
* The bag file will be written to ${REPO_ROOT}/data.bag