ORB-SLAM代码详解之代码框架

  • 工具篇
  • ORB-SLAM的代码框架

转载请注明出处:http://blog.csdn.net/c602273091/article/details/54933404

工具篇

在看代码方面,有不少软件推荐。我觉得首推还是得Understand【1】,这个工具可以把代码的流程图、依赖关系、整个框架给描述出来。有钱的话,买正版。其次的话,我推荐SourceInsight。这个工具比较方便看出你的类的定义。我在使用的时候一般是上面一半是代码,下面一半看类的定义。这样看代码就是事半功倍。

ORB-SLAM的代码框架

ORB-SLAM代码详解之代码框架_第1张图片
在这里我主要分析的是单目SLAM中的kitti数据集的流程图。

接下来从main函数开始抽丝剥茧:

/**
* This file is part of ORB-SLAM2.
*
* Copyright (C) 2014-2016 Ra煤l Mur-Artal  (University of Zaragoza)
* For more information see 
*
* ORB-SLAM2 is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* ORB-SLAM2 is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with ORB-SLAM2. If not, see .
*/


#include
#include
#include
#include
#include

#include

#include"System.h"

using namespace std;

// 从文件夹中load图片进来
void LoadImages(const string &strSequence, vector<string> &vstrImageFilenames,
                vector<double> &vTimestamps);

int main(int argc, char **argv)
{
    // 判断参数是否符合规范
    // ./mono_kitti path_to_vocabulary path_to_settings path_to_sequence
    // 一共有四个参数:可执行文件argv[0]、字典的目录argv[1]、配置的路径argv[2]、图片序列的路径argv[3]
    if(argc != 4)
    {
        cerr << endl << "Usage: ./mono_kitti path_to_vocabulary path_to_settings path_to_sequence" << endl;
        return 1;
    }

    // Retrieve paths to images
    vector<string> vstrImageFilenames;
    vector<double> vTimestamps;
    // 加载图片,传入图片路径、传回的是图片的文件名、每幅图片的时间戳
    LoadImages(string(argv[3]), vstrImageFilenames, vTimestamps);

    // 图片的个数
    int nImages = vstrImageFilenames.size();

    // 对SLAM系统进行初始化,传入字典和配置的路径
    // Create SLAM system. It initializes all system threads and gets ready to process frames.
    ORB_SLAM2::System SLAM(argv[1],argv[2],ORB_SLAM2::System::MONOCULAR,true);


    // Vector for tracking time statistics
    vector<float> vTimesTrack;
    // 计算追踪所花的时间
    vTimesTrack.resize(nImages);

    cout << endl << "-------" << endl;
    cout << "Start processing sequence ..." << endl;
    cout << "Images in the sequence: " << nImages << endl << endl;

    // Main loop
    cv::Mat im;
    for(int ni=0; ni// Read image from file
        im = cv::imread(vstrImageFilenames[ni],CV_LOAD_IMAGE_UNCHANGED);
        double tframe = vTimestamps[ni];

        if(im.empty())
        {
            cerr << endl << "Failed to load image at: " << vstrImageFilenames[ni] << endl;
            return 1;
        }

#ifdef COMPILEDWITHC11
        // 如果编译器可以编译c++11
        // 获取当前时间
        std::chrono::steady_clock::time_point t1 = std::chrono::steady_clock::now();
#else
        std::chrono::monotonic_clock::time_point t1 = std::chrono::monotonic_clock::now();
#endif

        // Pass the image to the SLAM system
        // 图片放入SLAM系统中进行追踪
        SLAM.TrackMonocular(im,tframe);

#ifdef COMPILEDWITHC11
        std::chrono::steady_clock::time_point t2 = std::chrono::steady_clock::now();
#else
        std::chrono::monotonic_clock::time_point t2 = std::chrono::monotonic_clock::now();
#endif

        double ttrack= std::chrono::duration_cast<std::chrono::duration<double> >(t2 - t1).count();
        // 计算追踪该图片花的时间
        vTimesTrack[ni]=ttrack;

        // Wait to load the next frame
        // 计算下一帧图片时间戳与当前时间戳的差值T,与追踪所需时间进行比较
        // 如果有必要就将当前线程暂停sleep
        // 主要是为了模拟时间情况,因为追踪结束以后下一帧可能还没来
        double T=0;
        if(ni1)
            T = vTimestamps[ni+1]-tframe;
        else if(ni>0)
            T = tframe-vTimestamps[ni-1];

        if(ttrack1e6);
    }

    // Stop all threads
    // 追踪完所有图片序列以后,关掉当前的线程
    SLAM.Shutdown();

    // Tracking time statistics
    // 对追踪的部分进行一个统计
    // 计算中位数,总数、平均值
    sort(vTimesTrack.begin(),vTimesTrack.end());
    float totaltime = 0;
    for(int ni=0; nicout << "-------" << endl << endl;
    cout << "median tracking time: " << vTimesTrack[nImages/2] << endl;
    cout << "mean tracking time: " << totaltime/nImages << endl;

    // Save camera trajectory
    // 保存整个相机的位姿的轨迹
    SLAM.SaveKeyFrameTrajectoryTUM("KeyFrameTrajectory.txt");

    return 0;
}

void LoadImages(const string &strPathToSequence, vector<string> &vstrImageFilenames, vector<double> &vTimestamps)
{
    ifstream fTimes;
    // 读4541幅图片的时间戳(kitti的00图片序列)
    string strPathTimeFile = strPathToSequence + "/times.txt";
    // string类型转化为char*类型
    fTimes.open(strPathTimeFile.c_str());
    while(!fTimes.eof())
    {
        string s;
        getline(fTimes,s);
        if(!s.empty())
        {
            // string类型转化为double类型
            // 使用stringstream类型
            stringstream ss;
            ss << s;
            double t;
            ss >> t;
            vTimestamps.push_back(t);
        }
    }
    // 使用image_0目录下的文件,这是双目摄像头的左边的摄像头的序列
    string strPrefixLeft = strPathToSequence + "/image_0/";

    const int nTimes = vTimestamps.size();
    vstrImageFilenames.resize(nTimes);

    for(int i=0; istringstream ss;
        // 设置0~nTime-1的图片的路径
        ss << setfill('0') << setw(6) << i;
        vstrImageFilenames[i] = strPrefixLeft + ss.str() + ".png";
    }
}

整个main函数看下来,真正有价值的地方在于:

  • SLAM系统初始化:ORB_SLAM2::System
  • SLAM(argv[1],argv[2],ORB_SLAM2::System::MONOCULAR,true);
  • SLAM系统的追踪部分:SLAM.TrackMonocular(im,tframe);

次要的地方是

  • 线程关闭:SLAM.Shutdown();
  • 保存整条轨迹: SLAM.SaveKeyFrameTrajectoryTUM(“KeyFrameTrajectory.txt”); 看了源码我觉得如果是kitti的数据集的话,应该调用SaveTrajectoryKITTI

当然在看代码之前把ORB-SLAM的论文看一遍非常重要。对于这片论文的介绍,可以看我的系列文章【2】【3】【4】。

参考链接:
【1】understand下载: http://www.qqtn.com/down/91669.html
【2】http://blog.csdn.net/c602273091/article/details/54348202
【3】http://blog.csdn.net/c602273091/article/details/54411989
【4】http://blog.csdn.net/c602273091/article/details/54428693

你可能感兴趣的:(RL,&,DL,&,SLAM,Autonomous,Driving)