sadmansk/minotaur-cpp

C++ implementation of minotaur - UWNRG controls program

Minotaur C++ Reimplementation

master	develop

Introduction

Minotaur is the controls software for our microrobots. It is being actively developed with the following goals in mind:

• Cross-platform development and deployment
• Maintainable object-oriented design and architecture
• Abstracted implementation and approachable interface
• High-speed architecture for real-time data crunching (read image recognition)
• Expose controller API that listens to Python scripts for quick navigation deployment

Please checkout the Projects tab for information about roadmap and current progress.

Setting up Minotaur on Windows

These will describe the steps for setting up an environment on Windows. This has been tested on a clean installation of Windows 10.

1. Download and install Git https://git-scm.com/download/win
2. Download and install Python 3.6 https://www.python.org/downloads
   • Select Add Python 3.6 to PATH
   • Under Customize installation ensure Add Python to environment variables and Precompiled standard library are checked
3. Download and install Qt https://www1.qt.io/download-open-source
   • Choose only Qt 5.6 for MinGW 4.9.2 32 bit
   • Qt ships with g++.exe and mingw32-make.exe
4. Download and install CMake https://cmake.org/download
   • Make sure to check Add CMake to the system PATH
5. Add the Qt tools directory to the System PATH: C:/Qt/tools/mingw492_32/bin
   • Make sure to restart PowerShell or cmd
6. Add the Qt 5.6 binaries to the System PATH: C:/Qt/5.6/mingw49_32/bin
7. Clone the repo git clone https://github.com/uwnrg/minotaur-cpp.git
   • cd minotaur-cpp
   • mkdir build
   • cd build
8. The following variables are needed by cmake
   • CMAKE_PREFIX_PATH points to Qt: C:/Qt/5.6/mingw49_32/lib/cmake
   • PYTHON_INCLUDE_DIRS points to Python headers: C:/Python36-32/include
   • PYTHON_LIBRARY points to Python library: C:/Python36-32/libs/python36.lib

Building Minotaur

9. cmake .. -G "MinGW Makefiles" -DCMAKE_PREFIX_PATH="C:/Qt/5.6/mingw49_32/lib/cmake" -DPYTHON_INCLUDE_DIRS="C:/Python36-32/include" -DPYTHON_LIBRARY="C:/Python36-32/libs/python36.lib"
10. mingw32-make
11. Double-click minotaur-cpp.exe

Post-setup

1. You can add the Python include and lib directories to your System PATH or
2. If you are using CLion, you can add CMAKE_PREFIX_PATH, PYTHON_INCLUDE_DIRS and PYTHON_LIBRARY as either environment variables or cmake flags
3. If CLion can't find g++ you can point it directly to the location in Build, Execution, and Deployment
4. You can also set the QTDIR environment variable to C:/Qt/5.6
5. You should now be able to build and run with CLion or cmake .. -G "MinGW Makefiles" and mingw32-make

Setting up Minotaur on MacOS

These will describe the basic steps for setting up an environment on MacOS. There may be issues if multiple versions of Python are installed on the system.

Installing Qt

1. Download the Qt community edition and install a version at least 5.6
2. I recommend 5.6 or 5.7 since the later versions take up much more space

Installing Python

1. You can either download Python 3.6 from the website and install or
2. Use brew install python3
3. Verify the python version (minimum 3.4) with python3 --version

Installing OpenCV 3

1. OpenCV 3 is the required version
2. Build and install with brew install opencv
3. Additional steps or dependencies (such as Python 2 and Python 3) listed here

Building Minotaur

Here you can either install CLion and import the project (recommended) or

1. git clone https://github.com/uwnrg/minotaur-cpp.git
2. cd minotaur-cpp
3. mkdir build
4. cd build
5. cmake ../ -DCMAKE_PREFIX_PATH=/path/to/qt/%VERSION%/clang_64/lib/cmake
6. make

And then run with ./minotaur-cpp. Note that if you're using CLion, go to Preference -> Build, Execution, Deployment -> CMake and add the same CMAKE_PREFIX_PATH argument.

CMake might have a hard time finding the correct Python version because Apple. Ping me @jeffniu22 if you've got any issues.

Setting up Minotaur on Ubuntu 16.04

From a fresh install, you will need these packages

sudo apt install build-essential cmake python3 python3-dev qt5-default libudev-dev libopencv-dev

You will need to build OpenCV 3 from source. Download the OpenCV 3 source and build. It is recommended to build with USE_QT=ON instead of GTK. Additional instructions found here.

Alternatively, if your computer is running amd64 architecture with Ubuntu 16.04, a prebuilt Debian package can be found here. Note that additional dependencies may be required.

Then clone the repository and build with

git clone https://github.com/uwnrg/minotaur-cpp.git
cd minotaur-cpp
mkdir build
cd build
cmake ../
make

And then run with ./minotaur-cpp

If Qt is unable to detect cameras, make sure to run

sudo apt install libqt5multimedia5-plugins

Using the GOTURN pretrained model

The default tracker uses the MIL model, and may be configured to use other tracker models. However, using the GOTURN model requires downloading and adding to the working directory the architecture descriptor file goturn.prototxt and the pretrained model data goturn.caffemodel. The latter is about 350 MB. These files are hosted here. Once acquired, add them to the working directory of the minotaur-cpp binary or in the CMakeLists.txt directory.

Building with Debug output off

Configure the CMake project with cmake -DNO_DEBUG=ON ...

Contributing

Please refer to the Contributing Guidelines.

License

Minotaur, Copyright (C) 2016 University of Waterloo Nano-Robotics Group

This program 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.

This program 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.