Skip to content

Home

Jump to bottom
Xu Liu edited this page Nov 2, 2024 · 1 revision

Welcome to the SLIDE_SLAM wiki!

Run RGBD raw bags detailed instructions

Each of these commands in each mentioned bullet point should be run in a separate terminal window. For each of these commands, make sure to source your workspace first.

  • First, start a roscore
roscore
  • Then set up sim time flag to true
rosparam set /use_sim_time true
  • Run Instance Segmentation

This is the front-end instance segmentation network which will run a YOLOv8 model to detect objects in the RGB images.

roslaunch object_modeller rgb_segmentation_f250.launch
  • Run Sync Semantics Node

It will launch the message sychronization node to sychronize the modelled objects and odom topics

roslaunch object_modeller sync_semantic_measurements.launch robot_name:=robot0 odom_topic:=/dragonfly67/quadrotor_ukf/control_odom
  • Run Process Cloud Node

This node is responsible for fitting object models to the segmented point clouds

roslaunch scan2shape_launch process_cloud_node_rgbd_indoor_with_ns.launch
  • Run Factor Graph SLAM Backend
roslaunch sloam single_robot_sloam_test.launch
  • Play The Bag File

Replace the YOUR_BAG_FILE with the name of the bag file you want to use.

rosbag play YOUR_BAG_FILE --clock -r 0.5 --topics /dragonfly67/quadrotor_ukf/control_odom /camera/aligned_depth_to_color/image_raw /camera/color/image_raw /camera/depth/image_rect_raw /camera/aligned_depth_to_color/image_raw:=/robot0/camera/aligned_depth_to_color/image_raw /camera/color/image_raw:=/robot0/camera/color/image_raw /camera/depth/image_rect_raw:=/robot0/camera/depth/image_rect_raw

Run LiDAR raw bags detailed instructions

Each of these commands in each mentioned bullet point should be run in a separate terminal window. For each of these commands, make sure to source your workspace first.

  • First, start a roscore
roscore
  • Then set up sim time flag to true
rosparam set /use_sim_time true
  • Run Faster-LIO with LiDAR Drivers

This will run the Faster-LIO algorithm with the LiDAR drivers to process the LiDAR and IMU packets and provide odometry along with a undisorted point cloud.

roslaunch scan2shape_launch run_flio_with_driver.launch
  • Run Semantic Segmentation

This is the front-end semantic segmentation network which will run a RangeNet++ model to segment objects in the LiDAR point clouds.

Make sure you edit the infer_node_params.yaml file present inside the scan2shape_launch/config folder and set the value of model_dir param to point to the path to the RangeNet++ model you downloaded in the previous step. Make sure to compelte the path with the / at the end.

roslaunch scan2shape_launch infer_node.launch
  • Run Sync Semantics Node

It will launch the message sychronization node to sychronize the modelled objects and odom topics

roslaunch object_modeller sync_semantic_measurements.launch robot_name:=robot0 odom_topic:=/Odometry
  • Run Process Cloud Node

This node is responsible for fitting object models to the segmented point clouds

roslaunch scan2shape_launch process_cloud_node_outdoor_with_ns.launch
  • Run Factor Graph SLAM Backend
roslaunch sloam single_robot_sloam_test_LiDAR.launch
  • Play The Bag File

Replace the YOUR_BAG_FILE with the name of the bag file you want to use.

rosbag play YOUR_BAG_FILE --clock --topics /os_node/lidar_packets /os_node/imu_packets /os_node/metadata
Clone this wiki locally