Skip to content

Installing Carmen LCAD on Ubuntu 14.04

Jump to bottom
Vinícius Brito Cardoso edited this page Nov 20, 2018 · 3 revisions

These are the instructions for a fresh install of CARMEN LCAD. It is expected that you already have minimal experience with Linux and know how to use the Terminal.

What do you need before starting?

  • A machine running Ubuntu 14.04.5 LTS. Other versions may work, but are not guaranteed. You can find it here. We advise against using a virtual machine because of the performance.
  • Patience. This is a long process and requires attention while following the steps.

Installation process

Initial Downloads

Update apt:

sudo apt-get update

Install the packages of subversion, gimp, meld, vim and git:

sudo apt-get install gimp meld subversion vim git

Download the next two items while doing the next steps.

Download Carmen through git :

git clone https://github.com/LCAD-UFES/carmen_lcad.git

Download MAE through git:

git clone https://github.com/LCAD-UFES/MAE.git

Install the dependencies

Install this packages required by Carmen:

sudo apt-get install swig \
libgtk2.0-dev \
qt-sdk \
libimlib2 libimlib2-dev \
imagemagick libmagick++-dev \
libwrap0 libwrap0-dev tcpd \
libncurses5 libncurses5-dev \
libgsl0-dev libgsl0ldbl \
libdc1394-22 libdc1394-22-dev libdc1394-utils \
cmake \
libgtkglext1 libgtkglext1-dev \
libgtkglextmm-x11-1.2-0 libgtkglextmm-x11-1.2-dev \
libglade2-0 libglade2-dev \
freeglut3 freeglut3-dev \
libcurl3 libcurl3-nss libcurl4-nss-dev \
libglew1.5 libglew1.5-dev libglewmx1.5 libglewmx1.5-dev libglew-dev \
libkml0 libkml-dev \
liburiparser1 liburiparser-dev \
libusb-1.0-0 libusb-1.0-0-dev libusb-dev \
libxi-dev libxi6 \
libxmu-dev libxmu6 \
build-essential libforms-dev \
byacc \
flex \
doxygen \
libespeak-dev libfftw3-dev \
libpng++

In the case an error occur because of libcheese-gtk23, install this packages and try again:

sudo apt-get install libglew-dev libcheese7 libcheese-gtk23 libclutter-gst-2.0-0 libcogl15 libclutter-gtk-1.0-0 libclutter-1.0-0

Install Java

sudo add-apt-repository ppa:webupd8team/java
sudo apt-get update && sudo apt-get install oracle-java8-installer
update-alternatives --display java

Edit the file /etc/environment:

sudo gedit /etc/environment

Add to the end of the file:

JAVA_HOME=/usr/lib/jvm/java-8-oracle

Install Eclipse

Download Eclipse from:

https://www.eclipse.org/downloads/eclipse-packages/

It's recommended to utilize the Eclipse IDE for C/C++ Developers. The current version of Eclipse is eclipse-cpp-mars-1-linux-gtk-x86_64.tar.gz

Unzip eclipse:

cd Downloads/
sudo mv eclipse-cpp-mars-1-linux-gtk-x86_64.tar.gz  /opt
cd /opt/
sudo tar -xvf eclipse-cpp-mars-1-linux-gtk-x86_64.tar.gz

Create a desktop link e edit it in /usr/share/applications:

sudo gedit /usr/share/applications/eclipse.desktop

Put the content below in file :

[Desktop Entry]
Name=Eclipse
Type=Application
Exec=/opt/eclipse/eclipse
Terminal=false
Icon=/opt/eclipse/icon.xpm
Comment=Integrated Development Environment
NoDisplay=false
Categories=Development;IDE
Name[en]=Eclipse

Install the dependencies of PCL:

sudo apt-get install libeigen3-dev libboost-all-dev libflann-dev libvtk5-dev cmake-gui

Install OpenCV 3.1:

Start installing the dependencies:

sudo apt-get install build-essential cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev \
python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libjasper-dev libdc1394-22-dev

Download and select the correct version:

sudo mkdir /opt/opencv3.1.0/
cd /opt/opencv3.1.0/
sudo git clone https://github.com/Itseez/opencv.git
sudo git clone https://github.com/Itseez/opencv_contrib.git
cd opencv
sudo git checkout 3.1.0
cd /opt/opencv3.1.0/opencv_contrib
sudo git checkout 3.1.0
cd /opt/opencv3.1.0/opencv

Build it:

sudo mkdir build
cd build
sudo cmake -D -DWITH_IPP=ON -D WITH_CUDA=OFF -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local \
-D OPENCV_EXTRA_MODULES_PATH=/opt/opencv3.1.0/opencv_contrib/modules /opt/opencv3.1.0/opencv/ ..
sudo make -j8
sudo make install

It is common to have problems with ippicv using this version of OpenCV. To fix it:

  • Edit /usr/local/lib/pkgconfig/opencv.pc
  • Add libdir3rd=${exec_prefix}/share/OpenCV/3rdparty/lib after libdir
  • Add -L${libdir3rd} at Libs

Install bullet, FANN and linux-CAN:

Download and extract:

sudo su
cd /usr/local/
wget https://storage.googleapis.com/google-code-archive-downloads/v2/code.google.com/bullet/bullet-2.78-r2387.tgz
wget http://downloads.sourceforge.net/project/fann/fann/2.2.0/FANN-2.2.0-Source.tar.gz
wget http://www.kvaser.com/software/7330130980754/V5_3_0/linuxcan.tar.gz
tar -xvf bullet-2.78-r2387.tgz 
tar -xvf linuxcan.tar.gz
tar -xvf FANN-2.2.0-Source.tar.gz
mv bullet-2.78 bullet

Build:

cd bullet
./configure
make
make install
cd ..
cd linuxcan
make
make install
cd ..
cd FANN-2.2.0-Source
mkdir build
cd build
cmake ..
make
make install

Configure the paths of installed packages

Edit the file /etc/ld.so.conf.d/opencv.conf:

gedit /etc/ld.so.conf.d/opencv.conf

Add to it's end:

/usr/local/lib

Execute:

ldconfig

edit /etc/bash.bashrc:

gedit /etc/bash.bashrc

Add to it's end:

PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig
export PKG_CONFIG_PATH

Exit root:

exit

Open the file ~/.bashrc:

gedit ~/.bashrc

Place in yours the content below:

#CARMEN
export PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/lib:/usr/lib/x86_64-linux-gnu:/usr/lib/i386-linux-gnu/:/usr/lib/libkml
export CARMEN_HOME=~/carmen_lcad

#OpenJaus
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$CARMEN_HOME/sharedlib/OpenJAUS/libopenJaus/lib:$CARMEN_HOME/sharedlib/OpenJAUS/libjaus/lib:$CARMEN_HOME/sharedlib/OpenJAUS/ojTorc/lib:$CARMEN_HOME/sharedlib/OpenJAUS/ojIARASim/lib

#Darknet
export DARKNET_HOME=$CARMEN_HOME/sharedlib/darknet
export LD_LIBRARY_PATH=$DARKNET_HOME/lib:$LD_LIBRARY_PATH

#MAE
export MAEHOME=~/MAE
export PATH=$PATH:$MAEHOME/bin

Install imlib and flycapture:

cd $CARMEN_HOME/ubuntu_packages/
sudo dpkg -i imlib_1.9.15-20_amd64.deb 
sudo dpkg -i imlib-devel_1.9.15-20_amd64.deb
tar -xvf flycapture2-2.5.3.4-amd64-pkg.tgz
cd flycapture2-2.5.3.4-amd64/
sudo apt-get install libglademm-2.4-1c2a libglademm-2.4-dev libgtkmm-2.4-dev libudev-dev
sudo sh install_flycapture.sh

Make some links:

sudo ln -s /usr/lib64/libgdk_imlib.so.1.9.15 /usr/lib64/libgdk_imlib.a
sudo ln -s /usr/src/linux-headers-3.8.0-30/ /usr/src/linux

Install PCL:

sudo add-apt-repository ppa:v-launchpad-jochen-sprickerhof-de/pcl
sudo apt-get update
sudo apt-get install libpcl-all

Install Kinect Camera:

sudo su
cd /usr/local
wget http://sourceforge.net/projects/libusb/files/libusb-1.0/libusb-1.0.19/libusb-1.0.19.tar.bz2
tar xvf libusb-1.0.19.tar.bz2
cd libusb-1.0.19
./configure
make
make install

if a error occur during installation, execute:

sudo apt-get install libudev-dev

Install TPLIB:

mkdir /usr/local/tplib
cd /usr/local/tplib
git clone git://github.com/OpenKinect/libfreenect.git
cd libfreenect
mkdir build
cd build
cmake ..
cp src/libfreenect.pc /usr/local/tplib/
make
cp ../src/libfreenect.pc.in src/libfreenect.pc 
cp ../fakenect/fakenect.sh.in fakenect/fakenect.sh
make install
ldconfig /usr/local/lib64/
exit

Execute:

glview

If give error, try:

freenect-glview

If give error, execute:

sudo ldconfig /usr/local/lib

Install library G2O:

sudo apt-get install cmake libsuitesparse-dev libqt4-dev qt4-qmake
cd /usr/local/
sudo git clone https://github.com/LCAD-UFES/g2o.git
cd /usr/local/g2o/trunk/build/
sudo cmake ../ -DBUILD_CSPARSE=ON -DG2O_BUILD_DEPRECATED_TYPES=ON -DG2O_BUILD_LINKED_APPS=ON -DG2O_BUILD_EXAMPLES=OFF
sudo make
sudo make install

Create the link of boost:

sudo ln -s /usr/lib/x86_64-linux-gnu/libboost_thread.so /usr/lib/x86_64-linux-gnu/libboost_thread-mt.so

Install MAE:

First, it's dependencies:

sudo apt-get install make g++ freeglut3-dev byacc libforms-dev libtiff4-dev libXi-dev libXmu-dev doxygen tcsh flex libdc1394-22-dev

Now, MAE itself:

  • Enter MAE directory:
cd $MAEHOME
  • Compile MAE:
make
  • Check if $MAEHOME/lib/libnet_conn.a was created
  • Check if $MAEHOME/bin/netcomp was created

Install Dlib:

cd /usr/local
sudo su
git clone https://github.com/davisking/dlib.git
cd dlib/
git checkout v19.4 -b dlib_v19.4
mkdir build
cd build/
In the IARA's computers, compile without CUDA:
cmake -D DLIB_USE_CUDA=OFF ..
If you need use CUDA, just to:
cmake .. 
make
make install

Install libwnn

cd /usr/local
git clone https://github.com/filipemtz/libwnn.git
cd libwnn
mkdir build
cd build
cmake ..
make -j 4
make install
exit

If cmake failed to find OpenCV, remove the generated files from the build directory, and run cmake with the option -DOpenCV_DIR=[path to directory build of your opencv installation]. Ex.:

cd /usr/local/libwnn/build
rm -rf *
cmake -DOpenCV_DIR=/opt/opencv3.1.0/opencv/build/ ..
make -j 8
make install
exit

Install cuDNN

Note: Install cuDNN library only if your computer has a NVIDIA GPU. First install the proper NVIDIA driver, then the NVIDIA CUDA Toolkit 7.5 or 8.0. Version 7.5 does not support NVIDIA Pascal architecture (GTX 1060, GTX 1070, GTX 1080, Titan X). For these GPU models use the CUDA Toolkit 8.0.

If using CUDA Toolkit 7.5 download cuDNN files by clicking here: cudnn-7.5-linux-x64-v5.1.tgz

cd ~/Downloads
tar -xzvf cudnn-7.5-linux-x64-v5.1.tgz
sudo cp -rH cuda/* /usr/local/cuda-7.5/
rm -rf cuda

If using CUDA Toolkit 8.0 download cuDNN files by clicking here: cudnn-8.0-linux-x64-v5.1.tgz

cd ~/Downloads
tar -xzvf cudnn-8.0-linux-x64-v5.1.tgz
sudo cp -rH cuda/* /usr/local/cuda-8.0/
rm -rf cuda

Make sure ~/.bashrc contains the following:

#CUDA and cuDNN
export PATH=/usr/local/cuda/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH
export CUDA_LIBS=/usr/local/cuda/lib64

Install Caffe

Note: Caffe may be installed even if your computer does not have NVIDIA GPU, CUDA Toolkit, cuDNN library. In this case Caffe will be installed with the CPU_ONLY flag activated.

sudo apt-get install libgoogle-glog-dev protobuf-compiler libhdf5-dev liblmdb-dev libleveldb-dev libsnappy-dev libatlas-base-dev
cd $CARMEN_HOME/sharedlib/ENet/caffe-enet
mkdir build && cd build
cmake ..
make all -j 8
make pycaffe

If cmake fails to find OpenCV, remove the generated files from the build directory, and run cmake with the option -DOpenCV_DIR=[path to directory build of your opencv installation]. For example:

cd $CARMEN_HOME/sharedlib/ENet/caffe-enet/build
rm -rf *
cmake -DOpenCV_DIR=/opt/opencv3.1.0/opencv/build/ ..
make -j 8
make pycaffe
sudo apt-get install python-matplotlib python-numpy python-pil python-scipy build-essential cython python-skimage python-protobuf

If cmake fails, you may compile with make, uncommenting the following line in the Makefile.config. Please refer to the generic Caffe installation guide for further help.

WITH_PYTHON_LAYER := 1

Please add the following to your ~/.bashrc file:

#Caffe ENet
export CAFFE_ENET_HOME=$CARMEN_HOME/sharedlib/ENet/caffe-enet
export CAFFE_HOME=$CAFFE_ENET_HOME
export PYTHONPATH=$CAFFE_ENET_HOME/python
export LD_LIBRARY_PATH=$CAFFE_ENET_HOME/build/lib:$LD_LIBRARY_PATH

In order to suppress the verbose messages of libcaffe, please add the following to your ~/.bashrc file:

export GLOG_minloglevel=1

Check the existence of file $CAFFE_ENET_HOME/include/caffe/proto/caffe.pb.h. If not present, please do the following:

cd $CAFFE_ENET_HOME
protoc src/caffe/proto/caffe.proto --cpp_out=.
mkdir include/caffe/proto
mv src/caffe/proto/caffe.pb.h include/caffe/proto/

Install CARMEN

Close all terminals and execute:

cd $CARMEN_HOME/src
./configure --nojava --nocuda --nozlib
  Should the C++ tools be installed for CARMEN: [Y/n] Y
  Should Python Bindings be installed: [y/N] N
  Searching for Python2.4... Should the old laser server be used instead of the new one: [y/N] N
  Install path [/usr/local/]: 
  Robot numbers [*]: 1,2

Before compiling CARMEN, we need to compile the module tracker beforehand for navigator_spline to work:

cd $CARMEN_HOME/src/tracker
make

It will give an error compiling, but it's okay.

To compile CARMEN, run:

cd $CARMEN_HOME/src
make

If give error of libusb.h, edit the file /usr/local/include/libfreenect.hpp:

sudo vim /usr/local/include/libfreenect.hpp

And alter the line #include <libusb.h> to:

#include <libusb-1.0/libusb.h>

If an error occur because of z_stream, try:

cd $CARMEN_HOME/src
make clean
./configure --nojava --nocuda
  Should the C++ tools be installed for CARMEN: [Y/n] Y
  Should Python Bindings be installed: [y/N] N
  Searching for Python2.4... Should the old laser server be used instead of the new one: [y/N] N
  Install path [/usr/local/]: 
  Robot numbers [*]: 1,2
make
Clone this wiki locally