Skip to content

Latest commit

 

History

History
 
 

introduction

Week 1: software installation and sample agent

Python

First you have to setup python, then install python libraries. The recommended way to install python libraries is using pip.

Ubuntu

sudo apt-get install python-pip
sudo pip install numpy matplotlib ipython

Windows

  1. download and install python: https://www.python.org/download/windows/
  2. install pip: https://pip.pypa.io/en/latest/installing.html
  3. install numpy and matplotlib
python -m pip install numpy matplotlib ipython

Learn Python

If you are new to python, you can try it out online, and follow Introduction to Python. If you want a little more depth, Python Tutorial is a great place to start, We also recommend to Learn Python the Hard Way.

You can try ipython notebooks that I used in the lecture by starting ipython notebook in this folder.

SimSpark

SimSpark is a generic simulator for various multiagent simulations. It supports developing physical simulations for AI and robotics research with an open-source application framework. We use customized version which has NAO V4.

Ubuntu

  1. add PPA and install
sudo add-apt-repository ppa:xu-informatik/simspark-spl
sudo apt-get update
sudo apt-get install rcssserver3d-spl
  1. start simspark in console:
simspark

Windows

  1. download zip package
  2. extract the zip package
  3. execute rcssserver3d.exe to start simspark

From source code (Other operation systems)

  1. download source code from https://github.com/xuyuan/SimSpark-SPL
  2. follow the instruction in http://simspark.sourceforge.net/wiki/index.php/Main_Page

Try out sample agent

start sample agent

  1. start simspark

  2. start a console (cmd in windows)

  3. go to the introduction source code folder

cd introduction
  1. start sample agent:
python spark_agent.py

Now, the spark_agent is connected to simspark, and you can see a robot in the simulation, but it does nothing. Let's program it.

Programming exercise

Open the following python files, write code follow comments in files.