Git Product home page Git Product logo

line_detector's Introduction

line_detector

A ros package for detecting lines and queues, and getting a position at the line's end.

To learn more, please refer to EyalSeg/line_detector/wiki.

Before After

Dependencies

  • ubuntu 16.04 LTS
  • ROS kinect
  • CUDA 9.*
  • python 2.7.*
  • python 3.5.* or above

The following python3 packges are required:

  • tensorflow 1.*
  • Keras
  • numpy
  • scikit-image
  • pycocotools
  • pip 9.0.1 or above

The following python2 packges are required:

  • OpenCV
  • scipy
  • numpy

installation

First of all, follow RoboTiCan's installation tutorial in order to install armadillo2 software.

Now in order to make armadillo's software compatible with this project, please open the file ~/catkin_ws/src/armadillo/armadillo2_utils/iai_kinect2/kinect2_registration/CMakeLists.txt and add this line add_definitions( -fexceptions ) at the end of the file. You may open a new terminal and use this command:

grep -qxF 'add_definitions( -fexceptions )' $HOME/catkin_ws/src/armadillo/armadillo2_utils/iai_kinect2/kinect2_registration/CMakeLists.txt || echo -e "\nadd_definitions( -fexceptions )" >> $HOME/catkin_ws/src/armadillo/armadillo2_utils/iai_kinect2/kinect2_registration/CMakeLists.txt

Then download and install an Image Detection Server. In this project we have used an implementation over Mask RCNN which you can clone and install from here. To do so, follow the instructions below.

Use the following commands in the terminal:

mkdir ~/catkin_ws/src/line_detection
cd ~/catkin_ws/src/line_detection
git clone https://github.com/bguplp/depthCamera.git
cd depthCamera/Mask_RCNN-master
python -m pip install -r requirements_py2.txt
python3 -m pip install -r requirements.txt
python3 setup.py install --user
sudo apt-get install python3-tk

Make sure you have tensorflow 1.*, you can check your current tensorflow version by running this command in the terminal:

For tensorflow 1.*

python3 -c 'import tensorflow as tf; print("tensorflow version:", tf.VERSION)'

For tensorflow 2.*

python3 -c 'import tensorflow as tf; print("tensorflow version:", tf.__version__)'

Now clone line_detector package into your ~/catkin_ws/src/line_detection and compile it, like that:

cd ~/catkin_ws/src/line_detection
git clone https://github.com/bguplp/line_detector.git
cd ~/catkin_ws
source ~/catkin_ws/devel/setup.bash
catkin_make

We have used 3D Gazebo human Models (3DGEMS) from here. You can use the following commands to download the folder we need to your ~/.gazebo/models/ directory:

wget  -P ~/Downloads/ "http://data.nvision2.eecs.yorku.ca/3DGEMS/data/miscellaneous.tar.gz"
cd ~/.gazebo/models
tar -xvzf ~/Downloads/miscellaneous.tar.gz 
cd miscellaneous
cp -a human_female_1 human_female_1_1 human_female_2 human_female_3 human_female_4 human_male_1 human_male_1_1 human_male_2 human_male_3 human_male_4 ~/.gazebo/models/
cd ..
rm -rv miscellaneous

Runing

In order to run line detector, use the following commands in the following order:

python2 http server communication

Open new terminal and run detection_server.py like this:

cd ~/catkin_ws/src/line_detection/depthCamera/src
python3 detection_server.py

For the gazebo simulation, use this command in new terminal:

source ~/catkin_ws/devel/setup.bash
roslaunch armadillo2 armadillo2.launch gazebo:=true kinect:=true world_name:="`rospack find line_detector`/coffee_line.world" map:="`rospack find line_detector`/coffee_line.yaml" have_map:=true move_base:=true amcl:=true lidar:=true x:="1.0" y:="1.0" Y:="1.0"

For the real armadillo2 robot, use the following command in new terminal:

roslaunch armadillo2 armadillo2.launch kinect:=true move_base:=true gmapping:=true lidar:=true

Launch line_end_detection.launch in new terminal with the folowing comand:

source ~/catkin_ws/devel/setup.bash
roslaunch line_detector line_end_detection.launch

After all the nodes are running and the required queue is in the camera frame, open new terminal and call the line_end_detection ros service with the folowing comand:

source ~/catkin_ws/devel/setup.bash
rosservice call /line_end_detection {}

python3 ROS communication

For gazebo simulation, use this command in new terminal:

source ~/catkin_ws/devel/setup.bash
roslaunch armadillo2 armadillo2.launch gazebo:=true kinect:=true world_name:="`rospack find line_detector`/coffee_line.world" map:="`rospack find line_detector`/coffee_line.yaml" have_map:=true move_base:=true amcl:=true lidar:=true x:="1.0" y:="1.0" Y:="1.0"

For real armadillo2 robot, use the following command in new terminal:

roslaunch armadillo2 armadillo2.launch kinect:=true move_base:=true gmapping:=true lidar:=true

Launch line_end_detection.launch in new terminal with the folowing comand:

source ~/catkin_ws/devel/setup.bash
roslaunch line_detector line_end_detection_py3.launch

After all the nodes are running and the required queue is in the camera frame, open new terminal and call the line_end_detection ros service with the folowing comand:

source ~/catkin_ws/devel/setup.bash
rosservice call /line_end_detection {}

Troubleshoot

  • I recommend to use this bash script to install nvidia driver, cuda, cudnn. It can solve lots of problems for you.

  • pip3 upgrade after installation by apt (ubuntu file system)

sudo apt install python3-pip
sudo pip3 install --upgrade pip setuptools
sudo apt update&& sudo apt upgrade python-pip
  • ImportError: No module named '_tkinter', please install the python3-tk package
sudo apt-get install python3.5-tk
  • ImportError: /opt/ros/kinetic/lib/python2.7/dist-packages/cv2.so: undefined symbol: PyCObject_Type
cd /opt/ros/kinetic/lib/python2.7/dist-packages/
sudo mv cv2.so cv2_ros.so
  • In case which you experience problem with running tensorflow on the gpu but you succeed to run it on the cpu you can try this approach. Install Cudnn, my recommended installation is here. Then do:
sudo update-initramfs -u

and reboot If that's haven't fix the problem, you probably should check your cuda version and consider replacing it to cuda-9.*. Or you can try that approach:

sudo ln -s /usr/local/cuda-<current-number>/lib64/libcudart.so.<current-number> /usr/local/cuda-<current-number>/lib64/libcudart.so.<correct-number>
sudo ln -s /usr/local/cuda-<current-number>/lib64/libcufft.so.<current-number> /usr/local/cuda-<current-number>/lib64/libcufft.so.<correct-number>
sudo ln -s /usr/local/cuda-<current-number>/lib64/libcurand.so.<current-number> /usr/local/cuda-<current-number>/lib64/libcurand.so.<correct-number>
sudo ln -s /usr/local/cuda-<current-number>/lib64/libcusolver.so.<current-number> /usr/local/cuda-<current-number>/lib64/libcusolver.so.<correct-number>
sudo ln -s /usr/local/cuda-<current-number>/lib64/libcusparse.so.<current-number> /usr/local/cuda-<current-number>/lib64/libcusparse.so.<correct-number>

line_detector's People

Contributors

talfeiner avatar eyalseg avatar bguplp avatar

Watchers

James Cloos avatar

Forkers

orhaimwerthaim

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.