Skip to content

dazory/diva2

Repository files navigation

diva2

자율주행차량의 실시간 다중센서데이터 수집·정제·분석 및 관제 통합 플랫폼 개발


📋 Description

항목 내용
프로젝트명 자율주행차량의 실시간 다중센서데이터 수집·정제·분석 및 관제 통합 플랫폼 개발
기간 2020.12.23 ~ 2021.06.18

To prepare for the unexpected driving environment that can occur when self-driving in a real road environment, a system is required to monitor, manage and process the vehicle's driving information from the outside. This paper is a study of remote streaming platforms that collect multi-sensor data from vehicles and communicate in real-time using ZeroMQ, Google Protocol Buffer. It visualizes data obtained from Camera, GPS, IMU, CAN, and LiDAR sensors in real-time using Qt, and provides playback function using time-synchronized data stored in the database. Also, it enables evaluation of algorithms using stored real-world driving data.


🚘 System Configuration

DIVA2의 전체 시스템 구성도는 다음과 같습니다.

system-configuration

  • Mobile Platform : 차량에 장착된 PC로, 센서 데이터 취득 및 처리를 담당
  • Server : 센서데이터를 통합 관리
  • Ground Station : 차량 외부에 위치하며 실시간으로 차량 상태 관제 및 테스트를 담당

✨ Experiments

✨ Specifications

실험이 진행된 시스템 환경은 다음과 같습니다.

  1. Mobile Platform

    mobile-platform

    차량 현대 i30
    카메라 Logitech C920
    LiDAR Ouster 3D LiDAR OS1
    IMU Microstrain 3DM-GX5-25
    GPS Ublox C94 M8P
    CAN i30 내장
    PC CPU : Intel® core™ i7-7820HK
    CPU @2.90GHz
    RAM : 8G
    GPU : NVIDIA® GeForce® GTX 1070
  2. Server

    • 3.5GHz의 인텔 코어 i7 프로세서

    • 6개의 물리적 12개의 논리적 코어

    • 256GB의 메모리, 8GB 램

    • 네트워크 대역폭 : 1000Mbps (850Mbps)

  3. Ground Station

    • CPU : Intel i5-1135G7

    • RAM : 16GB DDR4

    • GPU : Intel Iris Xe Graphics

✨ Demo

Ground Station의 주요 기능은 다음과 같습니다.

  • Real-time Visualization : Mobile Platform으로부터 실시간으로 데이터를 수신하여 차량의 상태를 시각화
  • Playback : Server에 저장된 데이터를 수신하여 차량의 상태를 시각화
  • Algorithm Test : Server에 저장된 데이터를 이용하여 알고리즘 테스트
  1. Real-time Visualization

realtime-visualization ← 클릭시 유튜브로 이동

  1. Playback

playback ← 클릭시 유튜브로 이동

  1. Algorithm Test

algorithm-test ← 클릭시 유튜브로 이동


📁 Document Structure

├─ diva2
│  ├─ build
│  ├─ MobilePlatform 
│  │      ├─ Sensing
│  │      └─ Visualization
│  ├─ Server
│  │      ├─ DBStoring
│  │      └─ TimeSynchronization
│  ├─ GroundStation
│  │      ├─ Playback
│  │      ├─ Visualization
│  │      ├─ AlgorithmTest
│  │      └─ AlgorithmTest_GUI
│  ├─ protobuf
│  ├─ service
│  ├─ test
│  │
│  ├─ CMakeLists.txt
│  ├─ README.md
│  └─ main.cpp
│          

🚀 Usage

🚀Prerequisites

항목 내용
개발환경 Linux Ubuntu 18.04 LTS
GUI 개발 플랫폼 QT 5.12.5
라이브러리·API 등 QT 5.12.5, CMake 3.10.2, OpenCV 3.2.0, OpenGL 4.4, QTI-4.0.47, PCL 1.11.1, VTK 7.1, PostgreSQL 10.17, protobuf 3.17.3, tomcat*, ZeroMQ, mscl, libmscl , libusb 1.0, flann 1.8, eigen3, boost, qhull*, freeglut3, libpqxx, expect

🚀 Run

  1. Download and unzip diva2 from this repository : https://github.com/dazory/diva2

  2. cd diva2
    mkdir build && cd build
    cmake ..
    make
  3. Now, you can see the executable files in each directory ...

    • Mobile Platform

      • Sensing

        cd diva2/build/MobilePlatform/Sensing
        ./MP_Sensing
      • Visualization

        cd diva2/build/MobilePlatform/Visualization
        ./MP_visualization
    • Server

      • TimeSynchronization

        cd diva2/build/Server/TimeSynchronization
        ./timesynchronization
      • DBStoring

        cd diva2/build/Server/DBStoring
        ./dbstoring
    • Ground Station

      • Algorithm wiki 확인 바람

      • Visualization

        cd diva2/build/GroundStation/Visualization
        ./GS_Visualization

🤝 Contributing

img @ka-yeon Avatar Avatar
dazory ka-yeon airpod2 yuntreee

and Thanks to DIVA for the advance study.

📝 Contact

If you have any questions, please email below.

About

DIVA2 is an integrated platform for real-time multi-sensor data collection, refining, analysis, and control of autonomous vehicle.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Contributors 4

  •  
  •  
  •  
  •