Install Instructions

User Manual:

Open the PDF directly: View PDF PDF.
Page Count: 3

DownloadInstall Instructions
Open PDF In BrowserView PDF
Yu-Ting Chung

Visual SLAM Install Instructions
I.

Description
This project is to develop a SLAM algorithm running on Raspberry Pi with camera. The
SLAM algorithm does both mapping and localization (location and orientation) at the same
time using only a camera. In addition, colored landmarks can be recognized during
navigation.

II.

Environment
Linux Ubuntu 18.04
g++ 7.3.0
OpenCV 2.4.13.6 (Required at least 2.4.3)
Eigen 3.3.4 (Required at least 3.1.0)

III.

Dependencies
1. C++ compiler is prerequisite
2. Required development tools:
sudo apt-get install cmake
sudo apt-get install libpython2.7-dev
sudo apt-get install libblas-dev liblapack-dev libglew-dev

3. Required dependencies
a. OpenCV (Download)
Installing required packages:
sudo apt-get install build-essential
sudo apt-get install cmake git libgtk2.0-dev pkg-config
libavcodec-dev libavformat-dev libswscale-dev
sudo apt-get install python-dev python-numpy libtbb2
libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev
libdc1394-22-dev

Downloading OpenCV sources and installing it:
cd ~/opencv
mkdir release
cd release
cmake -D CMAKE_BUILD_TYPE=RELEASE –D
CMAKE_INSTALL_PREFIX=/usr/local ..
make
sudo make install

b. Eigen3
Installing Eigen3:
sudo apt-get install libeigen3-dev

c. Pangolin (Download)
Downloading Pangolin sources and installing it:
cd ~/Pangolin
mkdir build
cd build
cmake ..
cmake –build .

d. DBoW2 (in project libs)
Installing DBoW2:
cd DBoW2
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make

e. G2o (in project libs)
Installing g2o:
cd g2o
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make

Note: Some errors might be occurring due to different environments. Common errors fixing:
Ø Pangolin error – deprecated constants
=> Adjust code in ffmpeg.cpp as here
Ø DBoW2 error: 'stdint-gcc.h' file not found
=> Replace 'stdint-gcc.h with stdint.h (Reference)
Ø G2o error: ‘tr1/unordered_map’ file not found
=> Remove all the tr1 references (Reference)

IV. Visual SLAM
ORB-SLAM2 is a C++ Visual SLAM algorithm library used in this project.
Github link: https://github.com/raulmur/ORB_SLAM2
Downloading project sources from Github and building it:
cd config
tar -xf ORBvoc.txt.tar.gz
cd ..
mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
make

“main” file will be generated in build folder after building.
Note: Some errors might be occurring due to different environments. Common errors fixing:
Ø Error: ‘usleep’ was not declared in this scope

=> Include  in System.h (Reference)
Ø Error: static_assert failed "Allocator::value_type must be same type as value_type"
=> Replace const KeyFrame* with KeyFrame* const (Reference)
Ø Error: No rule to make target
=> Replace libDBoW2.so and libg2o.so with libDBow.dylib and libg2o.dylib (Reference)

V.

Camera Calibration

Camera Calibration is necessary before running the project

Building it:
cd camera-calibration
mkdir build
cd build
cmake ..
make

“camera-calibration” file will be generated in build folder after building.
Source Exif Data: 
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
Page Count                      : 3
Title                           : Microsoft Word - InstallInstructions.docx
Producer                        : Mac OS X 10.13.6 Quartz PDFContext
Creator                         : Word
Create Date                     : 2018:07:19 02:22:10Z
Modify Date                     : 2018:07:19 02:22:10Z
EXIF Metadata provided by EXIF.tools

Navigation menu