Semantic Jaguar

This package contains ROS nodes that are able to generate a labelled pointcloud and create a semantic octomap representation.

Semantic octomap

Semantic KITTI - Sequence 06

Acknowledgement

This work cannot be done without many open source projects. Special thanks to

• semantic_slam, used as a template for octomap generation.
• octomap, used as our map representation.
• pcl library, used for point cloud processing.

License

TODO

Overview

Dependencies

• octomap

Installation

Install dependencies

rosdep install semantic_jaguar

Make

cd <your_catkin_work_space>
catkin_make

Run

The node receiving pointcloud data is the semantics_gen.py script.

Run with Kitti

First ensure you have the dataset structured as below:

dataset/
    └── sequences/
        └── 00/
            ├── velodyne/
            │   ├── 000000.bin
            │   └── 000001.bin
            ├── labels/
            │   ├── 000000.label
            │   └── 000001.label
            └── poses.txt

The kitti labels contain 28 classes including classes distinguishing non-moving and moving objects:

TODO:

• Add config for kitti
• Add instructions
• Provide example folder

Run semantic_mapping

You can run the semantic_cloud node and the octomap_generator node:

roslaunch semantic_jaguar semantic_mapping.launch

This will also launch rviz for visualization.

You can then move around the camera and construct semantic map. Make sure SLAM is not losing itself.

If you are constructing a semantic map, you can toggle the display color between semantic color and rgb color by running

rosservice call toggle_use_semantic_color

Run with ros bag

TODO

Configuration

NOTE: This is still in progress.

You can change parameters for launch. Parameters are in ./semantic_jaguar/params folder.

Note that you can set octomap/tree_type and semantic_cloud/point_type to 0 to generate a map with rgb color without doing semantic segmantation.

Parameters for octomap_generator node (octomap_generator.yaml)

namespace octomap

• pointcloud_topic
  • Topic of input point cloud topic
• tree_type
  • OcTree type. 0 for ColorOcTree, 1 for SemanticsOcTree using max fusion (keep the most confident), 2 for SemanticsOcTree using bayesian fusion (fuse top 3 most confident semantic colors). See project report for details of fusion methods.
• world_frame_id
  • Frame id of world frame.
• resolution
  • Resolution of octomap, in meters.
• max_range
  • Maximum distance of a point from camera to be inserted into octomap, in meters.
• raycast_range
  • Maximum distance of a point from camera be perform raycasting to clear free space, in meters.
• clamping_thres_min
  • Octomap parameter, minimum octree node occupancy during update.
• clamping_thres_max
  • Octomap parameter, maximum octree node occupancy during update.
• occupancy_thres
  • Octomap parameter, octree node occupancy to be considered as occupied
• prob_hit
  • Octomap parameter, hitting probability of the sensor model.
• prob_miss
  • Octomap parameter, missing probability of the sensor model.
• save_path
  • Octomap saving path. (not tested)

Parameters for semantic_cloud node (semantic_cloud.yaml)

namespace semantic_cloud

• cloud_topic
  • Topic for output pointcloud.
• point_type
  • Point cloud type, should be same as octomap/tree_type. 0 for intensity point cloud, 2 for semantic point cloud including top 3 most confident semanic colors and their confidences, 1 for semantic including most confident semantic color and its confident.
• frame_id
  • Point cloud frame id.
• file_path
  • Path to the kitt semantic dataset folder.