open-rdc / mcl_with_landmark

mcl_with_landmark

Requirement

theta_simple_stitching
branch: melodic-devel
https://github.com/open-rdc/theta_simple_stitching

yolov5_pytorch_ros
branch: detect_landmark
https://github.com/open-rdc/yolov5_pytorch_ros

yolo_to_landmarks
branch: main
https://github.com/open-rdc/yolo_to_landmark

emcl
branch: mcl_with_landmark
https://github.com/open-rdc/emcl

orne_navigation
https://github.com/open-rdc/orne_navigation
branch: emcl_with_landmark

Execute

for Ubuntu1804

Execute roscore on docker

1. docker run -it --net host --name roscore ros:melodic

Capture images from Theta

2. Launch Theta in Live mode Press the power and camera buttons to start

3. Change the authority sudo chmod 666 /dev/bus/usb/001/*

4. roslaunch theta_simple_stitching theta.launch

5. Launch rviz

6. Select /image/mercator/Image to display the image

Execute yolo5_pytorch_ros

cd ~/catkin_ws/src/yolov5_pytorch_ros/docker
docker-compose up -d
docker exec -it yolo bash

mkdir -p catkin_ws/src
cd catkin_ws/src
catkin_init_workspace
git clone https://github.com/open-rdc/yolov5_pytorch_ros
cd ..
catkin_make
source ~/catkin_ws/devel/setup.bash
roslaunch yolov5_pytorch_ros landmark_detector.launch

7. Select /detections_image_topic/Image to display the image on rviz

8. rosrun yolo_to_landmark yolo_to_landmark.py

Simulator

9. roslaunch orne_bringup orne_alpha_sim.launch

10. roslaunch orne_navigation_executor nav_static_map.launch emcl:=true

rqt_graph