autowarefoundation / lucid_vision_driver Goto Github PK

Lucid Vision Camera ROS2 Driver with Arena SDK

License: Apache License 2.0

CMake 19.46% C++ 73.28% Python 7.26%

lucid_vision_driver's Introduction

arena_camera

arena_camera node publishes image data collected from Lucid Vision Labs Triton GigE cameras in /lucid_vision/camera_"X" /image_raw topic as a ROS message( sensor_msgs/Image). This node can connect to multiple camera devices discovered by Lucid Vision Labs' ArenaSDK. In order to use this node, you need to install Lucid Vision Labs' ArenaSDK.

Installation

This camera driver has been tested on ROS2 Galactic(Ubuntu 20.04) and ROS2 Humble(Ubuntu 22.04).

Download ArenaSDK from here.
Install ArenaSDK.
Before connecting to the camera you need to set your IP address, according to the documentation of Triton Cameras your IP address can be set to 169.254.0.1.
Clone this driver to <your_workspace>/src/.
Check the Serial number of devices and change the serial number values in parameter files. (Serial number writes on the box of the camera) The serial number is different for each product.

If you have only one camera work on this param file : <your_ws>/src/arena_camera/param/lucid_vision_camera.param.yml If you have more than one camera work on this param file : <your_ws>/src/arena_camera/param/multi_camera.param.yml
Build your code with the following command.

colcon build
Source the directory and run the executable with following command.

ros2 launch arena_camera test_node_container.launch.py

7.1 You can check whether data is flowing or not and what is the rate of it with following commands.

ros2 topic hz /lucid_vision/<your_camera>/image_raw
Open another terminal and run Rviz2 with following command. rviz2

8.1 Add the /lucid_vision/<your_camera>/image_raw topic to Display panel. With following "Add">"By topic" section.

8.2 Set your "Fixed Frame" as "lucid_vision".

8.3 Set your "Reliability Policy" to "Best Efford". (Best effort works in UDP, Reliable works in TCP/IP)

8.4 Be sure that the visibility button is checked.

Camera Settings

Camera settings can be made in two ways;

Use default device settings.
1. Change camera settings with LUCID's ArenaView on a Windows computer.
2. Load settings to the camera device.
3. Run this driver using use_default_device_settings parameter as a true.
Change camera setting with rqt_reconfigure.
1. Open the terminal and run rqt_reconfigure with the following command.
  
  ros2 run rqt_reconfigure rqt_reconfigure
2. Change your camera settings with rqt_reconfigure GUI. Choose your camera from the list and change your settings. Choose desired exposure, gain and gamma values. You can also change the FPS of the image. (You can change your settings with ROS2 parameters too. You can find the parameters in the param file.)
3. Dump your camera settings with the following command.
  
  ros2 param dump /arena_camera_node --output-dir <your_workspace>/src/arena_camera/param/
  
  Run the camera node with the new param file. ros2 run arena_camera arena_camera_node_exe --ros-args --params-file <your_workspace>/src/arena_camera/param/arena_camera_node.yaml

lucid_vision_driver's People

Contributors

Stargazers

Watchers

Forkers

ajunlonglive kaancolak voltonomous lionator bresilla lrmput leo-drive marioney mebasoglu robicarr-karsan anuraktrajmathur avgteam-operakitb epengc

lucid_vision_driver's Issues

About Image Rect Bandwidth

First of all, I did what was mentioned in the readme file and build the workspace correctly. I replaced my own calibration parameters with those in the test_full_res.yaml file. I visualized via Rviz2 and there was no problem. But when I watch the bandwidths of the topics, I realized that the rectified image use much bw. Is it normal?

Also upper is /image and bottom is /image_rect

Network setting issue

All my LiDARs and PCs are under 192.168.0.1/24 subnet, the default lucid camera IP is under 169.254.0.1/16 if i am not wrong. If i only got one PC to run both perception and Autoware.universe, how can I reroute the lucid camera IP under the 192.168.0.1 subnet?

ROS2 galactic colcon build error

Environment:

Docker
ROS2 galactic
ubuntu 20.04

Camera model: TRI032S-CC

Reproduce:

Install the ArenaSDK inside the Docker
git clone this project under workspace/src/
cd back to root of workpace
source /opt/ros/galactic/setup.bash
colcon build --symlink-install

Error messages:

Starting >>> arena_camera
--- stderr: arena_camera                             
In file included from /home/ros/lucid_camera_ws/src/lucid_vision_driver/arena_camera/src/arena_camera_node.cpp:20:
/home/ros/lucid_camera_ws/src/lucid_vision_driver/arena_camera/include/arena_camera/arena_camera_node.h:28:10: fatal error: camera_info_manager/camera_info_manager.hpp: No such file or directory
   28 | #include <camera_info_manager/camera_info_manager.hpp>
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
make[2]: *** [CMakeFiles/arena_camera_node.dir/build.make:63: CMakeFiles/arena_camera_node.dir/src/arena_camera_node.cpp.o] Error 1
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [CMakeFiles/Makefile2:134: CMakeFiles/arena_camera_node.dir/all] Error 2
make: *** [Makefile:141: all] Error 2
---
Failed   <<< arena_camera [4.42s, exited with code 2]

Summary: 0 packages finished [4.59s]
  1 package failed: arena_camera
  1 package had stderr output: arena_camera

On top of that question:
Inside the param file, camera_info_url: "file:///home/zeys/projects/main_autoware/autoware/src/sensor_component/external/arena_camera/config/camera0_info.yaml" is being called for camera param file. What is this file used for? How can I modify it to fit my camera?

process has died

I installed the package following instructions but I see the package name is lucid_vision_driver while readme file says another. However, I am now using correct package name in the launch file and got this error:

zillur@trc:~$ ros2 launch lucid_vision_driver test_node_container.launch.py 
[INFO] [launch]: All log files can be found below /home/zillur/.ros/log/2023-06-18-12-40-10-272274-trc-4375
[INFO] [launch]: Default logging verbosity is set to INFO
[INFO] [component_container-1]: process started with pid [4388]
[component_container-1] [INFO] [1687117210.670763303] [perception.object_detection.camera_node_right]: Load Library: /home/zillur/ros2_ws/install/lucid_vision_driver/lib/libarena_camera_node.so
[component_container-1] [INFO] [1687117210.751479257] [perception.object_detection.camera_node_right]: Found class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode>
[component_container-1] [INFO] [1687117210.751533363] [perception.object_detection.camera_node_right]: Instantiate class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode>
[component_container-1] Camera readed from yaml file. Camera Name:camera_1 Frame id:camera_top/camera_link Serial no:221202855 Pixel_format:bgr8 FPS:15
[component_container-1]  ~ArenaCamerasHandler()
[ERROR] [component_container-1]: process has died [pid 4388, exit code -11, cmd '/opt/ros/humble/lib/rclcpp_components/component_container --ros-args -r __node:=camera_node_right -r __ns:=/perception/object_detection'].
^C[WARNING] [launch]: user interrupted with ctrl-c (SIGINT)
[WARNING] [launch_ros.actions.load_composable_nodes]: Abandoning wait for the '/perception/object_detection/camera_node_right/_container/load_node' service response, due to shutdown.
zillur@trc:~$

Failed to Load multiple nodes/cameras.

I want to load 2 cameras simultaneously, but the second camera throws an error:

ros2 launch lucid_vision_driver test_node_container.launch.py [INFO] [launch]: All log files can be found below /home/vwp/.ros/log/2023-04-20-11-04-32-212034-datalynx-8259 [INFO] [launch]: Default logging verbosity is set to INFO [INFO] [component_container-1]: process started with pid [8271] [component_container-1] [INFO] [1681981472.596805394] [perception.object_detection.lucid_camera_node_container]: Load Library: /home/vwp/autoware/install/arena_camera/lib/libarena_camera_node.so [component_container-1] [INFO] [1681981472.694545831] [perception.object_detection.lucid_camera_node_container]: Found class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode> [component_container-1] [INFO] [1681981472.695266744] [perception.object_detection.lucid_camera_node_container]: Instantiate class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode> [component_container-1] Camera readed from yaml file. Camera Name:camera_1 Frame id:camera_top/camera_link Serial no:231301220 Pixel_format:bgr8 FPS:19 [component_container-1] [WARN] [1681981474.116701298] [ARENA_CAMERA_HANDLER]: Not possible to set exposure value when auto exposure is enabled. [component_container-1] [WARN] [1681981474.120789087] [ARENA_CAMERA_HANDLER]: Not possible to set gain value when auto gain is enabled. [component_container-1] Camera:1852729698 is created. [component_container-1] [INFO] [1681981474.126826927] [arena_camera_node_right]: camera calibration URL: file:///home/vwp/autoware/src/vwp/external/lucid_vision_driver/config/test.yaml [component_container-1] Camera idx:1852729698 acquisition thread. [component_container-1] [INFO] [1681981474.150741900] [ARENA_CAMERA]: Not possible to use hardware based binning for horizontal binning:4 , software binning will be used. [component_container-1] [INFO] [1681981474.167400934] [ARENA_CAMERA]: Not possible to use hardware based binning for vertical binning:4 , software binning will be used. [INFO] [launch_ros.actions.load_composable_nodes]: Loaded node '/arena_camera_node_right' in container '/perception/object_detection/lucid_camera_node_container' [component_container-1] [INFO] [1681981474.393411868] [perception.object_detection.lucid_camera_node_container]: Found class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode> [component_container-1] [INFO] [1681981474.393524752] [perception.object_detection.lucid_camera_node_container]: Instantiate class: rclcpp_components::NodeFactoryTemplate<ArenaCameraNode> [component_container-1] Camera readed from yaml file. Camera Name:camera_2 Frame id:camera_top/camera_link Serial no:231301251 Pixel_format:bgr8 FPS:19 [component_container-1] [ERROR] [1681981474.604538974] [perception.object_detection.lucid_camera_node_container]: Component constructor threw an exception [ERROR] [launch_ros.actions.load_composable_nodes]: Failed to load node 'arena_camera_node_left' of type 'ArenaCameraNode' in container '/perception/object_detection/lucid_camera_node_container': Component constructor threw an exception

Here from my launch file:

`def generate_launch_description():
launch_arguments = []

context = LaunchContext()
camera_param_path = os.path.join(
    FindPackageShare("arena_camera").perform(context),
    "param/test.param.yaml"
)
with open(camera_param_path, "r") as f:
    camera_yaml_param = yaml.safe_load(f)["/**"]["ros__parameters"]

nodes = []

nodes.append(
    ComposableNode(
        package="arena_camera",
        plugin="ArenaCameraNode",
        name="arena_camera_node_right",
        parameters=[{"camera_name": camera_yaml_param['camera_name'],
                     "frame_id": camera_yaml_param['frame_id'],
                     "pixel_format": camera_yaml_param['pixel_format'],
                     "serial_no": 231301220, #camera_yaml_param['serial_no'],
                     "camera_info_url": camera_yaml_param['camera_info_url'],
                     "fps": camera_yaml_param['fps'],
                     "horizontal_binning": camera_yaml_param['horizontal_binning'],
                     "vertical_binning": camera_yaml_param['vertical_binning'],
                     "use_default_device_settings": camera_yaml_param['use_default_device_settings'],
                     "exposure_auto": camera_yaml_param['exposure_auto'],
                     "exposure_target": camera_yaml_param['exposure_target'],
                     "gain_auto": camera_yaml_param['gain_auto'],
                     "gain_target": camera_yaml_param['gain_target'],
                     "gamma_target": camera_yaml_param['gamma_target'],
                     }],
        remappings=[
        ],
        extra_arguments=[
            {"use_intra_process_comms": True}
        ],
    )
)

nodes.append(
    ComposableNode(
        package="arena_camera",
        plugin="ArenaCameraNode",
        name="arena_camera_node_left",
        parameters=[{"camera_name": "camera_2",#camera_yaml_param['camera_name'],
                     "frame_id": "camera_top/camera_link",#camera_yaml_param['frame_id'],
                     "pixel_format": camera_yaml_param['pixel_format'],
                     "serial_no": 231301251, #camera_yaml_param['serial_no'],
                     "camera_info_url": camera_yaml_param['camera_info_url'],
                     "fps": camera_yaml_param['fps'],
                     "horizontal_binning": camera_yaml_param['horizontal_binning'],
                     "vertical_binning": camera_yaml_param['vertical_binning'],
                     "use_default_device_settings": camera_yaml_param['use_default_device_settings'],
                     "exposure_auto": camera_yaml_param['exposure_auto'],
                     "exposure_target": camera_yaml_param['exposure_target'],
                     "gain_auto": camera_yaml_param['gain_auto'],
                     "gain_target": camera_yaml_param['gain_target'],
                     "gamma_target": camera_yaml_param['gamma_target'],
                     }],
        remappings=[
        ],
        extra_arguments=[
            {"use_intra_process_comms": True}
        ],
    )
)

container = ComposableNodeContainer(
    name="lucid_camera_node_container",
    namespace="/perception/object_detection",
    package="rclcpp_components",
    executable="component_container",
    composable_node_descriptions=nodes,
    output="both",
)

return LaunchDescription(
    [
        *launch_arguments,
        container
    ]
)

Do i miss something here?

parameter folder

Hi,
I can't see any parameter folders on my installation.
Is the repo up to date? Or, we should create a new yml file for it?

If you have only one camera work on this param file : <your_ws>/src/arena_camera/param/lucid_vision_camera.param.yml If you have more than one camera work on this param file : <your_ws>/src/arena_camera/param/multi_camera.param.yml

Build error with iron

This issue is fairly old but it seems to be the root cause of a problem building this package with iron. The declare_parameter function behavior has changed slightly.

ros2/rclcpp#1691 (comment)
ros2/rclcpp#1522

Should be a relatively easy (and backwards compatible I think?) fix to add default values or add rclcpp::PARAMETER_STRING etc to the Camera Setting constructor to allow unset values.


/home/workspace/src/lucid_vision_driver/src/arena_camera_node.cpp:57:26: error: no matching function for call to ‘ArenaCameraNode::declare_parameter(const char [12])’
   57 |   this->declare_parameter("camera_name");
      |   ~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~

(and) 

In file included from /opt/ros/iron/include/camera_info_manager/camera_info_manager/camera_info_manager.hpp:45,
                 from /home/workspace/src/lucid_vision_driver/include/arena_camera/arena_camera_node.h:24,
                 from /home/workspace/src/lucid_vision_driver/src/arena_camera_node.cpp:17:
/opt/ros/iron/include/rclcpp/rclcpp/node.hpp:479:3: note: candidate: ‘template<class ParameterT> auto rclcpp::Node::declare_parameter(const string&, const ParameterT&, const ParameterDescriptor&, bool)’
  479 |   declare_parameter(
      |   ^~~~~~~~~~~~~~~~~
/opt/ros/iron/include/rclcpp/rclcpp/node.hpp:479:3: note:   template argument deduction/substitution failed:
/home/workspace/src/lucid_vision_driver/src/arena_camera_node.cpp:57:26: note:   candidate expects 4 arguments, 1 provided

autowarefoundation / lucid_vision_driver Goto Github PK

lucid_vision_driver's Introduction

Installation

Camera Settings

lucid_vision_driver's People

Contributors

Stargazers

Watchers

Forkers

lucid_vision_driver's Issues

About Image Rect Bandwidth

Network setting issue

ROS2 galactic colcon build error

process has died

Failed to Load multiple nodes/cameras.

parameter folder

Build error with iron

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent