Nvidia Jetson Agx Xavier 在Ros中调用GMSL2相机

一、背景

        在ros中调用GMSL2摄像头，刚开始是通过修改官方驱动包ros-meloidc-usb-cam，可能是修改的地方不对，一直报错，调用失败，要是有大佬修改成功，希望能交流一下。

        后来借鉴了一下别人关于这方面的博客，成功拿到了图像数据。理论上jetson orin/nanotx2等也可以用。

        这是我的包，可根据自己情况修改jetsongmls2相机驱动-编解码文档类资源-CSDN下载

二、环境

        1、设备 Nvidia Jetson Agx Xavier

        2、jetpack 4.6.1

        3、opencv 4.1.1

        4、ros-melodic

        5、python3.6.9

        6、GMSL2相机+相机采集板

三、驱动包环境搭建

1、编译python3版cv_bridge

        参考我的上篇博客，此步不可省略，因为opencv4.1.1是python3版本的，所以就得使用python3版本的cv_bridge，而官方默认版本为python2版，如不自己构建cv_bridge，必须安装python2版本的opencv版本，使用默认cv_bridge。

在ROS-meloidc中使用python3 cv_bridge_Ponnyao的博客-CSDN博客

2、创建ROS工作空间

mkdir -p ~/ros_camera_ws/src
cd ~/ros_camera_ws/
catkin_make
source devel/setup.bash

3、创建ROS功能包learning_image_transport 

cd ~/ros_camera_ws/src
catkin_create_pkg learning_image_transport roscpp std_msgs cv_bridge image_transport sensor_msgs 

4、编写发布订阅节点C++文件

        在功能包learning_image_transport下的src目录中建立两个cpp文件。

1、my_publisher.cpp

cd ~/ros_camera_ws/src/learning_image_transport/src/
gedit my_publisher.cpp

写入以下代码：

#include <ros/ros.h>  
#include <image_transport/image_transport.h>  
#include <opencv2/highgui/highgui.hpp>  
#include <cv_bridge/cv_bridge.h>  
#include <sstream> // for converting the command line parameter to integer  
 
int main(int argc, char** argv)  
{  
    // Check if video source has been passed as a parameter  
    if(argv[1] == NULL)   
    {  
        ROS_INFO("argv[1]=NULL\n");  
        return 1;  
    }  
 
    ros::init(argc, argv, "image_publisher");  
    ros::NodeHandle nh;  
    image_transport::ImageTransport it(nh);  
    image_transport::Publisher pub = it.advertise("camera/image", 1);  
 
    // Convert the passed as command line parameter index for the video device to an integer  
    std::istringstream video_sourceCmd(argv[1]);  
    int video_source;  
    // Check if it is indeed a number  
    if(!(video_sourceCmd >> video_source))   
    {  
        ROS_INFO("video_sourceCmd is %d\n",video_source);  
        return 1;  
    }  
 
    cv::VideoCapture cap(video_source);  
    // Check if video device can be opened with the given index  
    if(!cap.isOpened())   
    {  
        ROS_INFO("can not opencv video device\n");  
        return 1;  
    }  
    cv::Mat frame;  
    sensor_msgs::ImagePtr msg;  
 
    ros::Rate loop_rate(5);  
    while (nh.ok()) 
    {  
        cap >> frame;  
        // Check if grabbed frame is actually full with some content  
        if(!frame.empty()) 
        {  
            msg = cv_bridge::CvImage(std_msgs::Header(), "bgr8", frame).toImageMsg();  
            pub.publish(msg);  
            //cv::Wait(1);  
    	}  
    }
    
    ros::spinOnce();  
    loop_rate.sleep();  
}  

2、my_subscriber.cpp

gedit my_subscriber.cpp

写入以下代码：

#include <ros/ros.h>  
#include <image_transport/image_transport.h>  
#include <opencv2/highgui/highgui.hpp>  
#include <cv_bridge/cv_bridge.h>  
 
void imageCallback(const sensor_msgs::ImageConstPtr& msg)  
{  
	try  
	{  
		cv::imshow("view", cv_bridge::toCvShare(msg, "bgr8")->image); 
		// cv::waitKey(30);  
	}  
	catch (cv_bridge::Exception& e)  
	{  
		ROS_ERROR("Could not convert from '%s' to 'bgr8'.", msg->encoding.c_str());  
	}  
}  
 
int main(int argc, char **argv)  
{  
	ros::init(argc, argv, "image_listener");  
	ros::NodeHandle nh;  
	cv::namedWindow("view");  
	cv::startWindowThread();  
	image_transport::ImageTransport it(nh);  
	image_transport::Subscriber sub = it.subscribe("camera/image", 1,imageCallback);  
	ros::spin();  
	cv::destroyWindow("view");  
}  

5、配置文件修改

cd ~/ros_camera_ws/src/learning_image_transport/
gedit CMakeLists.txt

填入以下内容：

find_package(OpenCV REQUIRED)  
# add the publisher example  
add_executable(my_publisher src/my_publisher.cpp)  
 
target_link_libraries(my_publisher ${catkin_LIBRARIES} ${OpenCV_LIBRARIES})  
 
# add the subscriber example  
add_executable(my_subscriber src/my_subscriber.cpp)  
target_link_libraries(my_subscriber ${catkin_LIBRARIES} ${OpenCV_LIBRARIES}) 

6、编译ROS包

cd ~/ros_camera_ws/
catkin_make  

7、测试

1、启动ros核心：

roscore

2、启动发布节点：

cd ~/ros_camera_ws/
source devel/setup.bash
rosrun learning_image_transport my_publisher 0 

0代表默认摄像头,如果有多个摄像头,则第二个是1,依次类推。

3、启动订阅节点：

cd ~/ros_camera_ws/
source devel/setup.bash
rosrun learning_image_transport my_subscriber

8、其他节点订阅图像

在对应功能包scripts文件夹下新建image_view.py

#!/usr/bin/env python3
#coding:utf-8
 
import rospy
import cv2
from sensor_msgs.msg import Image
from cv_bridge import CvBridge,CvBridgeError
 
def callback(data):         	
	cv_image=CvBridge().imgmsg_to_cv2(data,"bgr8")
	cv2.imshow("view",cv_image)
	cv2.waitKey(1)
 
if __name__ == '__main__':
    rospy.init_node('image_acquistion')
    rospy.Subscriber('camera/image', Image, callback)
    rospy.spin()
    cv2.destoryAllWindows()

rosrun 你的包名 image_view.py

参考博客

JETSON AGX XAVIER GMSL2接口相机驱动_努力划水的博客-CSDN博客_gmsl2接口