1、速腾16线激光测试参考：

源码地址：

https://github.com/RoboSense-LiDAR/ros_rslidar

多线激光使用、建图总结见：

https://blog.csdn.net/xingdou520/article/details/85098314

2、测试目的：

1、测试速腾聚创ROS驱动程序的执行过程，数据流向；

2、确认pub的数据格式，时间戳问题；

3、每一场数据包含的激光测量的具体角度问题确认。

4、点云格式确认，是否可以自定义点云格式，从而减小数据拷贝造成的时间、资源浪费。

3、测试结果：

1、rslidar_driver每次pub一场数据的时候，每场数据的起始角度都不一样。

2、每一场数据超过360度，在361-364度之间。

3、rslidar_driver驱动程序，每一场数据包含84个UDP数据包（rslidarPacket.msg每个数据包1248字节数据），每个包间隔1.1-1.4ms,合起来刚好100ms一场数据。每次读取84个完整数据包后pub一次数据（rslidarSacn.msg），

4、驱动程序发布数据的时间戳是激光传感器传上来的最后一包的数据的时候，ROS给数据添加的时间戳。

double time1 = ros::Time::now().toSec();UDP方式读取一包数据// Average the times at which we begin and end reading.  Use that to
// estimate when the scan occurred. Add the time offset.
double time2 = ros::Time::now().toSec();
pkt->stamp = ros::Time((time2 + time1) / 2.0 + time_offset);  //时间戳

5、rslidar_pointcloud包作用，订阅rslidarSacn.msg，并根据标定文件将原始数据转为sensor_msgs::PointCloud2数据发布出去。

/** @brief Callback for raw scan messages. 原始激光数据的回调函数*/void Convert::processScan( const rslidar_msgs::rslidarScan::ConstPtr &scanMsg){pcl::PointCloud<pcl::PointXYZI>::Ptr outPoints(new pcl::PointCloud<pcl::PointXYZI>);  //定义要发布的点云数据格式outPoints->header.stamp = pcl_conversions::toPCL(scanMsg->header).stamp;              //使用原始数据的时间戳outPoints->header.frame_id = scanMsg->header.frame_id;                                //使用源数据的frame_id// process each packet provided by the driverbool finish_packets_parse = false;for (size_t i = 0; i < scanMsg->packets.size(); ++i)                        //连续解析所有的数据包，速腾16线激光 每场数据84个数据包{if(i == (scanMsg->packets.size()-1)){//ROS_INFO_STREAM("Packets per scan: "<< scanMsg->packets.size());finish_packets_parse = true;}data_->unpack(scanMsg->packets[i], outPoints,finish_packets_parse);     //解析数据包，将原始数据转换为点云格式数据}sensor_msgs::PointCloud2 outMsg;      //定义ROS标准的点云消息pcl::toROSMsg(*outPoints, outMsg);    //将自定义的点云数据转换为ROS点云消息格式//if(outPoints->size()==0){//    ROS_INFO_STREAM("Height1: "<<outPoints->height<<" Width1: "<<outPoints->width);//}output_.publish(outMsg);              //发布点云消息}
}

6、可以用protobuf的方式代替sensor_msgs::PointCloud2，通过内存共享的方式实现多节点之间数据共享。

4、测试代码：

void print_time(struct tm *stm_)
{printf("T:%d-%02d-%02d %02d:%02d:%02d\n",stm_->tm_year ,stm_->tm_mon,stm_->tm_mday,stm_->tm_hour,stm_->tm_min,stm_->tm_sec);
}int stamp_test(uint8_t *data)
{int flag = 0;struct tm stm_;if ((data[0] == 0x55) && (data[1] == 0xaa) && (data[2] == 0x05) && (data[3] == 0x0a)){memset(&stm_, 0, sizeof(stm_));stm_.tm_year = (int)data[20] + 100;stm_.tm_mon  = (int)data[21] - 1;stm_.tm_mday = (int)data[22];stm_.tm_hour = (int)data[23];stm_.tm_min  = (int)data[24];stm_.tm_sec  = (int)data[25];double time_stamp = 0;time_stamp = 0.001 * (256 * data[26] + data[27]) + 0.000001 * (256 * data[28] + data[29]);static double time_stamp_lost = time_stamp;double stamp_double = mktime(&stm_) + time_stamp;static uint32_t seconds = ros::Time(stamp_double).sec;  //上一次 收到的时间 //if (ros::Time(stamp_double).sec - seconds > 0) //秒 发生变化 if(time_stamp_lost > time_stamp){                seconds = ros::Time(stamp_double).sec;std::cout << std::endl;print_time(&stm_);std::cout << std::setprecision(20) << "stamp_double:" << stamp_double << std::endl;    std::cout << std::setprecision(6);std::cout <<"time_stamp:" << time_stamp << "\tseconds: " << seconds ;int azimuth = 256 * data[44] + data[45];static int angle_lost = azimuth;// std::cout << " angle:\t" << azimuth /100.0 << "\tdiff:"<< (azimuth - angle_lost)/100.0 << std::endl;std::cout << " angle:\t" << azimuth /100.0 << "\terror:"<< (azimuth - 9000)/100.0 << std::endl;angle_lost = azimuth;flag= 1;}time_stamp_lost = time_stamp;}return flag;
}/** poll the device**  @returns true unless end of file reached*/bool rslidarDriver::poll(void) {// Allocate a new shared pointer for zero-copy sharing with other nodelets.rslidar_msgs::rslidarScanPtr scan(new rslidar_msgs::rslidarScan);scan->packets.resize(config_.npackets);// Since the rslidar delivers data at a very high rate, keep// reading and publishing scans as fast as possible.std::cout << " poll: " << std::endl;for (int i = 0; i < config_.npackets; i++) {  //84packetswhile (true) {// keep reading until full packet receivedint rc = input_->getPacket(&scan->packets[i], config_.time_offset);if (rc == 0) {uint8_t buf[50];for (size_t ii = 0; ii < 50; ii++){buf[ii] = scan->packets[i].data[ii];}// std::cout << " test "<< std::endl;if(stamp_test(buf)==1){std::cout << " packet: " << i << std::endl;}if(i == 0){uint8_t data[50];data[44] = scan->packets[i].data[44];data[45] = scan->packets[i].data[45];int azimuth = 256 * data[44] + data[45];// std::cout << " angle:\t" << azimuth /100.0 << "\tdiff:"<< (azimuth - angle_lost)/100.0 << std::endl;//std::cout << "packet num:" << i << " angle:\t" << azimuth /100.0 << std::endl;static int _angle_lost = azimuth;// std::cout << " angle:\t" << azimuth /100.0 << "\tdiff:"<< (azimuth - angle_lost)/100.0 << std::endl;std::cout << "packet num:" << i << " First Packet angle:\t" << azimuth /100.0 << "\tpub data anlge:"<< (azimuth+36000-_angle_lost)/100.0 << std::endl;_angle_lost = azimuth;}if(i == 83){uint8_t data[50];data[44] = scan->packets[i].data[44];data[45] = scan->packets[i].data[45];int azimuth = 256 * data[44] + data[45];// std::cout << " angle:\t" << azimuth /100.0 << "\tdiff:"<< (azimuth - angle_lost)/100.0 << std::endl;std::cout << "packet num:" << i << " Last Packet angle:\t" << azimuth /100.0 << std::endl;}break;       // got a full packet?}if (rc < 0) return false; // end of file reached?}}// publish message using time of last packet readROS_DEBUG("Publishing a full rslidar scan.");scan->header.stamp = scan->packets[config_.npackets - 1].stamp;scan->header.frame_id = config_.frame_id;output_.publish(scan);// notify diagnostics that a message has been published, updating its statusdiag_topic_->tick(scan->header.stamp);diagnostics_.update();// printf("npackets:%d\n",config_.npackets);return true;}

【激光雷达】速腾聚创16线激光雷达驱动程序测试总结相关推荐

速腾聚创16线激光雷达rslidar-16的ros驱动安装与rviz点云显示
原链接: https://community.bwbot.org/topic/520 运行测试平台:小强ROS机器人速腾聚创16线激光雷达rslidar-16的供电是直流12v 3A,通信接口是rs ...
速腾聚创16线激光雷达ros驱动安装与rviz点云展示说明
.速腾聚创16线激光雷达ros驱动安装与rviz点云展示说明 V1.0.0 – by Holden Date : 2021-10-12 文章目录 .速腾聚创16线激光雷达ros驱动安装与rviz点云展 ...
基于ros 在TX2上装速腾聚创16线激光雷达RS-LiDAR
准备工作: tx2开发板一块(ubantu16.04) 速腾激光雷达16线网线一块显示屏(用于连接tx2) 1.安装雷达驱动安装依赖 sudo apt-get install libpcap-d ...
速腾聚创16线激光雷达录的数据包运行LeGO-LOAM
1.run.launch文件修改如下: <launch><param name="/use_sim_time" v ...
速腾聚创Robosense16线激光雷达SLAM建图（一）————Hector建图
因为最近拿到一个Robosense16线的激光雷达,这是一个三维的激光雷达,所以就拿着这个雷达把一些开源的经典的SLAM算法都跑了一遍,中间也遇到了一些问题,所以就打算在博客上记录下来,供大家交流参考 ...
速腾聚创16线雷达配置
auto enp8s0 iface enp8s0 inet static address 10.7.5.88 netmask 255.0.0.0 gateway 10.7.5.100 因为松灵的厂家改 ...
【速腾聚创混合固态激光雷达RS-M1简介与实物开箱】
[RS-M1系列 - 0]速腾聚创混合固态激光雷达RS-M1简介与实物开箱这是我在CSDN的第一篇博客,文中难免有疏漏甚至错误,如有错误敬请在评论区指出,相互学习~ 网上关于传统机械式激光雷达的技术 ...
速腾聚创32线雷达雷达，RVIZ显示激光点云
1.下载驱动文件 ahren@ahren:~/ld_ws/src$ git clone https://github.com/RoboSense-LiDAR/ros_rslidar 2.在工作空间ld ...
法雷奥ScaLa核心人物加入：如何带领速腾聚创踏上车规激光雷达量产之路？
2017年7月,奥迪A8作为全球首款产量L3级自动驾驶车型一经亮相便引起行业内广泛讨论.而奥迪A8搭载的法雷奥四线激光雷达SacLa则是业内首款面向自动驾驶汽车的车规级激光雷达产品--后者在自动驾驶行 ...

【激光雷达】速腾聚创16线激光雷达驱动程序测试总结

1、速腾16线激光测试参考：

2、测试目的：

3、测试结果：

4、测试代码：

【激光雷达】速腾聚创16线激光雷达驱动程序测试总结相关推荐

最新文章

热门文章