5.2.2 Display

Web

Introduction

Web is used to preview camera images (JPEG format) and algorithm results. The images and algorithm results are transmitted to the PC browser through the network and rendered for display. The display interface also supports displaying only the video without rendering the intelligent results.

Code Repository: (https://github.com/D-Robotics/hobot_websocket)

Supported Platforms

Platform	System	Function
RDK X3, RDK X3 Module, RDK X5	Ubuntu 20.04 (Foxy), Ubuntu 22.04 (Humble)	Start MIPI camera human detection and display images and algorithm results through web

Preparation

RDK

Confirm that the camera F37 is correctly connected to the RDK.
Confirm that the PC can access the RDK through the network.
Confirm that TogetheROS.Bot has been successfully installed.

Usage

RDK

a. Launch mipi_cam

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

ros2 launch mipi_cam mipi_cam.launch.py mipi_video_device:=F37

b. Launch encoding

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

ros2 launch hobot_codec hobot_codec_encode.launch.py

c. Launch WebSocket

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

ros2 launch websocket websocket.launch.py websocket_image_topic:=/image_jpeg websocket_only_show_image:=true

Open a PC browser (Chrome/Firefox/Edge) and enter http://IP:8000 to view the image and algorithm effects. IP refers to the RDK IP address.

Notes

WebSocket uses port 8000. If the port is already in use, the launch will fail. Here are some solutions:
- Use the command lsof -i:8000 to check the processes that are occupying port 8000, and use kill <PID> to close the process, and then relaunch WebSocket.
- If the user does not want to stop the service that is currently using port 8000, you can modify the listen port number in the configuration file /opt/tros/lib/websocket/webservice/conf/nginx.conf to a port number that is greater than 1024 and not being used. After modifying the port number, the URL used in the browser also needs to be modified accordingly.

HDMI

Introduction

This chapter introduces the use of displaying camera nv12 images through HDMI. RDK can display real-time image effects by connecting to a monitor via HDMI, corresponding to the hobot_hdmi package.

Code Repository: (https://github.com/D-Robotics/hobot_hdmi)

Supported Platforms

Platform	System	Function
RDK X3, RDK X3 Module, RDK X5	Ubuntu 20.04 (Foxy), Ubuntu 22.04 (Humble)	Start MIPI camera and display image through HDMI

Preparation

RDK

RDK has been flashed with the Ubuntu 20.04/22.04 system image provided by D-Robotics.
TogetheROS.Bot has been successfully installed on the RDK.
RDK is connected to a monitor via HDMI.

Instructions

RDK

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

If use RDK X5, run commod:

# 关闭桌面显示
sudo systemctl stop lightdm
# 复制运行依赖
cp -r /opt/tros/${TROS_DISTRO}/lib/hobot_hdmi/config/ .

ros2 launch hobot_hdmi hobot_hdmi.launch.py device:=F37

Result Analysis

The following information is displayed in the running terminal:

[INFO] [launch]: All log files can be found below /root/.ros/log/2022-07-27-15-27-26-362299-ubuntu-13432
[INFO] [launch]: Default logging verbosity is set to INFO
[INFO] [mipi_cam-1]: process started with pid [13434]
[INFO] [hobot_hdmi-2]: process started with pid [13436]

The monitor displays the image as follows: hdmi

RViz2

Introduction

TogetheROS.Bot is compatible with ROS2 Foxy version. To conveniently preview image effects, you can use RViz2 to get images.

Supported Platforms

Platform	System	Sample Function
RDK X3, RDK X3 Module, RDK X5	Ubuntu 20.04 (Foxy), Ubuntu 22.04 (Humble)	Start the MIPI camera to capture images and use RViz2 to preview on PC

Preparation

RDK

RDK has flashed with the Ubuntu 20.04/22.04 system image provided by D-Robotics.
RDK has successfully installed tros.b.
The PC has installed Ubuntu 20.04, ROS2 Foxy Desktop version, and the data visualization tool RViz2. The PC and RDK are on the same network segment (the first three segments of the IP address are the same).

Reference for ROS2 Foxy installation: (https://docs.ros.org/en/foxy/Installation/Ubuntu-Install-Debians.html)

On the PC, install RViz2 with the command: sudo apt install ros-foxy-rviz-common ros-foxy-rviz-default-plugins ros-foxy-rviz2

Usage

RDK

SSH into the development board and start the corresponding program on the board

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

# Start the F37 camera to publish images in BGR8 format
ros2 launch mipi_cam mipi_cam.launch.py mipi_out_format:=bgr8 mipi_image_width:=480 mipi_image_height:=272 mipi_io_method:=ros mipi_video_device:=F37

Note: Do not change the mipi_out_format arbitrarily. RViz2 only supports image formats like RGB8, RGBA8, BGR8, BGRA8, etc.

If the following information is output, it means that the node has been successfully started

[INFO] [launch]: All log files can be found below /root/.ros/log/2022-08-19-03-53-54-778203-ubuntu-2881662
[INFO] [launch]: Default logging verbosity is set to INFO
[INFO] [mipi_cam-1]: process started with pid [2881781]

A new window is created in the RDK to execute the topic query command and the results are as follows:

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

# Query topics
ros2 topic list

Output:

/camera_info
/image_raw
/parameter_events
/rosout

On the PC, the current topics are queried using the following command and the results are as follows:

Foxy
Humble

source /opt/ros/foxy/setup.bash

source /opt/ros/humble/setup.bash

ros2 topic list

Output:

/camera_info
/image_raw
/parameter_events
/rosout

Subscribing to a topic and previewing camera data on the PC:

Foxy
Humble

source /opt/ros/foxy/setup.bash

source /opt/ros/humble/setup.bash

ros2 run rviz2 rviz2

On the RViz2 interface, first click the "add" button, then select the published image based on the topic, which in this example is named /image_raw. Then click "image":

rviz2-config

The image result is as follows:

rviz2-result

Attention

If the PC terminal's ros2 topic list does not recognize the camera topic, please check the following:
- Check if RDK is publishing images properly:
  - Foxy
  - Humble
  # Configure the tros.b environment source /opt/tros/setup.bash
  # Configure the tros.b environment source /opt/tros/humble/setup.bash
```
ros2 topic list
```
  Output:
```
/camera_info
/image_raw
/parameter_events
/rosout
```
- Check if the PC and RDK networks can ping each other;
- Check if the IP addresses of the PC and RDK have the same first three digits;

RQt

Introduction

TogetheROS.Bot is compatible with ROS2 Foxy and supports previewing compressed format images through RQt, greatly reducing network bandwidth consumption.

Supported Platforms

Platform	System	Function
RDK X3, RDK X3 Module, RDK X5	Ubuntu 20.04 (Foxy), Ubuntu 22.04 (Humble)	Start MIPI camera to capture images and use RQt to preview on PC

Preparation

RDK

RDK has been flashed with the provided Ubuntu 20.04/22.04 system image.
RDK has successfully installed tros.b.
PC has been installed with Ubuntu 20.04 system, ROS2 Foxy desktop version, and the visualization tool RQt. Both PC and RDK are on the same network segment (with the same first three digits of the IP addresses).

ROS2 Foxy installation reference: (https://docs.ros.org/en/foxy/Installation/Ubuntu-Install-Debians.html)

Installation method for rqt-image-view on PC terminal: sudo apt install ros-foxy-rqt-image-view ros-foxy-rqt

Usage

RDK

SSH into the development board and start relevant programs on the board:

a. Start F37 camera

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

ros2 launch mipi_cam mipi_cam.launch.py mipi_image_width:=640 mipi_image_height:=480 mipi_video_device:=F37

b. Start hobot_codec and publish compressed format images

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

ros2 launch hobot_codec hobot_codec_encode.launch.py codec_out_format:=jpeg codec_pub_topic:=/image_raw/compressed

If the program output shows the following information, it means the nodes have been successfully launched

[INFO] [launch]: All log files can be found below /root/.ros/log/2023-05-15-17-08-02-144621-ubuntu-4755
[INFO] [launch]: Default logging verbosity is set to INFO
[INFO] [mipi_cam-1]: process started with pid [4757]
[mipi_cam-1] This is version for optimizing camera timestamp 

[INFO] [launch]: All log files can be found below /root/.ros/log/2023-05-15-17-08-17-960398-ubuntu-4842
[INFO] [launch]: Default logging verbosity is set to INFO
[INFO] [hobot_codec_republish-1]: process started with pid [4844]

Subscribe to the topic on the PC and preview the camera data;

Foxy
Humble

source /opt/ros/foxy/setup.bash

source /opt/ros/humble/setup.bash

ros2 run rqt_image_view rqt_image_view

Select the topic /image_raw/compressed, and the image is as follows:

Notes

If ros2 topic list does not recognize the camera topic on the PC, perform the following troubleshooting steps:
- Check if the RDK is publishing images correctly
  - Foxy
  - Humble
  # Configure the tros.b environment source /opt/tros/setup.bash
  # Configure the tros.b environment source /opt/tros/humble/setup.bash
```
ros2 topic list
```
  Output:
```
/camera_info
/hbmem_img000b0c26001301040202012020122406
/image_raw
/image_raw/compressed
/parameter_events
/rosout
```
- Check if the PC and RDK can ping each other;
- Check if the PC and RDK have the same first three segments of IP address;

Foxglove

Introduction

Foxglove is an open-source toolkit that includes both online and offline versions. It aims to simplify the development and debugging of robotic systems. It provides a range of features for building robot applications.

In this section, we will primarily use the data recording and playback feature of Foxglove: Foxglove allows recording the data of ROS2 topics into files for subsequent playback and analysis. This is very useful for system troubleshooting, performance optimization, and algorithm debugging.

In the demonstration, we will use the hobot_visualization package developed by TogetheROS.Bot to convert intelligent inference results into ROS2 rendered topic information.

Code repository: (https://github.com/D-Robotics/hobot_visualization)

Supported Platforms

Platform	System	Function
RDK X3, RDK X3 Module, RDK X5	Ubuntu 20.04 (Foxy), Ubuntu 22.04 (Humble)	Offline object detection, and display images and algorithm effects using Foxglove

Preparation

RDK

Confirm that TogetheROS.Bot has been successfully installed.
Confirm that the PC can access the RDK via the network.

Usage

RDK

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

export CAM_TYPE=fb

ros2 launch hobot_visualization hobot_vis_render.launch.py

At the same time, log in to another terminal using SSH and record topic information on the board side:

Foxy
Humble

# Configure the tros.b environment
source /opt/tros/setup.bash

# Configure the tros.b environment
source /opt/tros/humble/setup.bash

# Record rosbag data, which will be generated in the current working directory
ros2 bag record -a

Play rosbag data on the Foxglove online page

In a PC browser (chrome/firefox/edge), enter (https://foxglove.dev/studio) to access the Foxglove website.

PS: Registration is required for first-time use. You can register using a Google account or a third-party email.
Enter the visualization function interface.
Click to select the local rosbag file.
Open the layout interface. In the top right corner of the layout interface, click on the settings, select the icon, and open the play marker rendering message function.
Click Play
View Data

Note

Foxglove visualizes image data using the official ROS2 message format and supports image encoding formats. For more details, please refer to (https://foxglove.dev/docs/studio/panels/image).
When recording messages with rosbag, it may record topic information from other devices. To ensure clean rosbag data, you can set export ROS_DOMAIN_ID=xxx, such as export ROS_DOMAIN_ID=1.

5.2.2 Display

Web​

Introduction​

Supported Platforms​

Preparation​

RDK​

Usage​

RDK​

Notes​

HDMI​

Introduction​

Supported Platforms​

Preparation​

RDK​

Instructions​

RDK​

Result Analysis​

RViz2​

Introduction​

Supported Platforms​

Preparation​

RDK​

Usage​

RDK​

Attention​

RQt​

Introduction​

Supported Platforms​

Preparation​

RDK​

Usage​

RDK​

Notes​

Foxglove​

Introduction​

Supported Platforms​

Preparation​

RDK​

Usage​

RDK​

Note​

Web

Introduction

Supported Platforms

Preparation

RDK

Usage

RDK

Notes

HDMI

Introduction

Supported Platforms

Preparation

RDK

Instructions

RDK

Result Analysis

RViz2

Introduction

Supported Platforms

Preparation

RDK

Usage

RDK

Attention

RQt

Introduction

Supported Platforms

Preparation

RDK

Usage

RDK

Notes

Foxglove

Introduction

Supported Platforms

Preparation

RDK

Usage

RDK

Note