AMOS Project 2016 - Group 5

Collision avoidance using object detection and inter-car communication

Summary:

This project will implement a scenario needed for autonomous driving with the help of software agents: A bus stops at a bus stop vis-à-vis to a school. A car that drives behind the bus cannot see the children starting to run across the street in front of the bus. Another car on the other side of the street detects the children and is able to warn the first car.

Main tasks:

• Implement a simple object detection algorithm
• Integrate a software agent library
• Implement the scenario using software agents and inter-ECU (electronic control unit, i.e. car) communication

Setup:

• Operating System: Linux Ubuntu (also tested with Mac OS X 10.10)
• SW Agents library: CAF: C++ Actor Framework
• Library for detection: OpenCV
• Messaging format: HDF5, Google Protobuf

Run project with docker:

Try our Docker image and run this project without installing all dependencies.
For installing Docker and pulling the image take a look at our wiki.
Use our scripts to start the programs.

Project dependencies:

• Min. CMake 2.8: Download, Install
• HDF5: Download & Install
• Google Protobuf 2.6.1: Download, Install
• Integrate OpenCV extra modules in OpenCV building: Download, Install do this before building OpenCV!
• OpenCV 3.1: Download, Install
• CAF: C++ Actor Framework 0.14.5: [Download & Install] (https://github.com/actor-framework/actor-framework)
• A [OpenMP Compiler] (http://openmp.org/wp/openmp-compilers/)

Clone project:

$ git clone https://github.com/JoHeinrich/amos-ss16-proj5.git
$ cd amos-ss16-proj5

Build project:

$ mkdir build
$ cd build
$ cmake ..
$ make

Run project on your computer:

Just the detection

$ ./amos-ss16-proj5 FULL/PATH/TO/VIDEO
e.g.:
$ ./amos-ss16-proj5 video.mp4

Detection and communication:

1. Start the server:
$ ./amos-ss16-proj5 PORT

e.g.:(using port 8080)
$ ./amos-ss16-proj5 8080

2. Start the detection (use another terminal window/tab)
$ ./amos-ss16-proj5 PORT SERVER_IP FULL/PATH/TO/VIDEO

e.g.: (using port 8080 and localhost)
$ ./amos-ss16-proj5 8080 127.0.0.1 video.mp4

Source code documentation:

See our Doxygen site