如果此开源项目所改进的内容对贵战队有帮助请Star本项目，若有使用本项目的能量机关模块或借鉴思路后的实车击打效果，希望能将视频发一份给作者本人（qq：1042277145），这对作者有非常大的帮助，十分感谢！

1. list the USB port, ensuring the camera is connected to USB bus port 3.0. (lsusb)
2. update the list of available software packages and upgrade them (sudo apt update && sudo apt upgrade)
3. More packages such as cmake to install (sudo apt install git cmake build-essential pkg-config)
4. Install dependencies for RealSense camera hardware communication, GUI, and graphic rendering(sudo apt install libusb-1.0-0-dev libgtk-3-dev libglfw3-dev libgl1-mesa-dev libglu1-mesa-dev)
5. Clone most recent realsense library and locate to that folder: (git clone https://github.com/IntelRealSense/librealsense.git cd librealsense)
6. Create a build directory and configure it makelist file so it can be compiled in local system ( mkdir build && cd build cmake ../ -DBUILD_EXAMPLES=true )
7. compile and install( make -j$(nproc) sudo make install )
8. test GUI( realsense-viewer )
9. Integrate with CV package.

1. pull docker ( sudo docker pull ghcr.io/fateryu/rm_vision:latest )

docker was installed by snap. So check docker status via: sudo systemctl status snap.docker.dockerd.service

2. configure the docker( docker run -it --name rv_devel
   --privileged --network host
   -v /dev:/dev -v $HOME/.ros:/root/.ros -v ws:/ros_ws
   ghcr.io/fateryu/rm_vision:latest
   ros2 launch foxglove_bridge foxglove_bridge_launch.xml )

Second time accessing the docker:( sudo docker ps -a: list dockers

!!! sudo docker start rv_devel: start docker named rv_devel

!!! sudo docker exec -it rv_devel bash: open interactive shell inside the container. You can access it as it is a separate machine

3. remove duplicate folder as ros_ws/rm_auto_aim and rows_ws/src/rm_auto_aim contain the same file( rm -rf /ros_ws/rm_auto_aim

4. install missing python package( sudo apt-get update sudo apt-get install python3-ament-package )

5. After you move back to the docker, source the ROS2 environment before building ensuring all dependencies are found:( source /opt/ros/humble/setup.bash )

6. build the workspace:( cd /ros_ws colcon build --symlink-install )

7. source ros2 and packages to current shell(docker): ( source /opt/ros/humble/setup.bash source /ros_ws/install/setup.bash )

8. run the armor_dectector package:( ros2 run armor_detector armor_detector_node )

Now you can get parameters.

1. Run realsense camera in your docker_bash_1: ( ros2 launch realsense2_camera rs_launch.py )

You run the camera in one docker and retrieve the data in that docker but with another opened bash.

1. open that docker in a new bash:( sudo docker start rv_devel sudo docker exec -it rv_devel bash )

2. source ros2 and packages: ( source /opt/ros/humble/setup.bash source /ros_ws/install/setup.bash )

3. retrieve the output RGB, depth data from bash 1:( ros2 topic list )

4. Remap the topic camera info to armor_detector node:( ros2 run armor_detector armor_detector_node --ros-args -r /image_raw:=/camera/camera/color/image_raw -r /camera_info:=/camera/camera/color/camera_info ) • Its local /image_raw topic to /camera/camera/color/image_raw • Its local /camera_info topic to /camera/camera/color/camera_info

5. Verify if the detector is receiving data from camera topic:( ros2 topic echo /detector/armors )

rm_vision 项目旨在为 RoboMaster 队伍提供一个规范、易用、鲁棒、高性能的视觉框架方案，为 RM 开源生态的建设添砖加瓦

如果本开源项目对于贵战队的视觉技术发展起到了实质性的帮助作用，请在机器人上贴上以下标签以助力该项目的推广，十分感激！

Origin Author: Chen Jun

Develop Author: Zheng Yu

• [rm_vision] 使用 LifeCycle 管理节点
• [armor] 识别器灯条识别由旋转外接矩形更换为最小二乘法
• [armor] 识别器调试话题加入角点信息，供后续数据录制使用
• [armor] 识别器灯条增加外接矩形中二值化占比过滤条件，减少将大块白光识别为灯条所占用资源
• [armor] 识别器增加任务触发条件，适配能量机关等多任务切换
• [armor] 跟踪器增加视野内多目标间切换功能
• [armor] 跟踪器扩展卡尔曼滤波过程噪声矩阵中增加平移与旋转参数负相关关系
• [armor] 可视化适配 Foxglove 新版本
• [armor] 控制端选板及开火控制逻辑
• [armor] 跟踪器扩展卡尔曼滤波运动模型由匀速模型改为匀加速模型
• [armor] 识别器角点PNP结合激光测距传感器自动调参
• [armor] 处理装甲板的灯条边缘受装甲板物理结构遮挡问题
• [armor] 预测效果评价器
• [serial] 通信适配能量机关
• [serial] 通信适配自瞄切换目标
• [serial] 通信新增控制端时间戳（计算通信与运算延时）
• [buff] 新增能量机关自动瞄准算法模块
• [buff] 优化能量机关扇叶神经网络识别角点效果（准确率、帧率）
• [buff] 优化跟踪器在大能量机关任务下的状态机逻辑
• [buff] 控制端开火控制
• [record] 新增自动录包模块
• [record] 与相机节点使用同一进程
• [record] 控制端新增物理按键启停录制

[changed] 装甲板自动瞄准算法模块 https://github.com/FaterYU/rm_auto_aim

[new] 能量机关自动瞄准算法模块 https://github.com/FaterYU/rm_buff

MindVision 相机模块 https://github.com/FaterYU/ros2_mindvision_camera

HikVision 相机模块 https://github.com/FaterYU/ros2_hik_camera

机器人云台描述文件 https://github.com/FaterYU/rm_gimbal_description

[changed] 串口通讯模块 https://github.com/FaterYU/rm_serial_driver

视觉算法仿真器 https://github.com/FaterYU/rm_vision_simulator

[new] 自动录包模块 https://github.com/FaterYU/rm_auto_record

拉取镜像

docker pull ghcr.io/fateryu/rm_vision:latest

构建开发容器

docker run -it --name rv_devel \
--privileged --network host \
-v /dev:/dev -v $HOME/.ros:/root/.ros -v ws:/ros_ws \
ghcr.io/fateryu/rm_vision:latest \
ros2 launch foxglove_bridge foxglove_bridge_launch.xml

构建运行容器

docker run -it --name rv_runtime \
--privileged --network host --restart always \
-v /dev:/dev -v $HOME/.ros:/root/.ros -v ws:/ros_ws \
ghcr.io/fateryu/rm_vision:latest \
ros2 launch rm_vision_bringup vision_bringup.launch.py

TBD