Maze Runner

First project in Digital Business Line Training at Obechar 2023. The project aims to support everyone to get used to with popular tools in software development, a basic workflow of a website and SCRUM.

Prerequisites

Before working on the project, you need to have a good understanding about git. Belows are the things you need to do first:

Fork this repo to your team's repository.
Learn to use basic commands of git including git clone, git add, git commit, git push.
Learn the basic workflow of a git (especialy common branches and their usages) and supported commands for this. The commands are git checkout, git merge, git rebase, git branch.
Learn to track and control your commits with git status, git reset --hard, git reset --soft, git diff.

Next, download an IDE to support your coding. The recommended is VSCode. If you select VSCode, try to install some popular extensions to support your development with Python and git.

Let's start your first step with Maze Runner

In Maze Runner, your team will try to develop a bot which try to find the shortest path in a maze to collect coins. At the end of the project, we will host a battle between two bots of two teams to find the winner. Try your best!

maze_metadata.json

The map of the maze is respresented in a simple .json file. This file contains the metadata (well, if you don't know what is this, search it!) and from this file, all other components can understand and interact with the maze. The format of the file is like this example:

{
    "width": 10,
    "height": 10,
    "obstacles": [[0, 1], [1, 1], [2, 1]],
    "bot": [5, 8],
    "coin": [0, 0]
}

Here, the width and height represent the size of the map, obstacles is a list of coordinate (row_index, column_index) of cells that are occluded, bot is the position (like obstacles) of the bot, and coin is the position (the same with bot) of the coin.

So, from the above example, a maze may looks like:

************
*o*        *
* *        *
* *        *
*          *
*          *
*        X *
*          *
*          *
*          *
*          *
************

REMEMBER, this is just a simple example. The real one is very challenging!

bot.py

Let's develop bot.py which represents as your bot. What it does is to read the maze_metadata.json and write the next action to a file named action.txt iteratively. The command line to run bot.py in terminal must follow this:

python bot.py -i maze_metadata.json -o action.txt

Here, the -i is an argument which defines the input file, and -o defines the output file.

The bot must only send 1 action for each step, after only when the position of the bot in maze_metadata.json is changed. The action must be either left, right, up or down. The output action is represented by an action.txt file. For example, with the given metadata, the bot might plan a path to go left, up, left, down, right and so on. The action.txt in this case will be like:

  left
  up
  left
  down
  right

Each line is equivalent to an action at each step.

Try your best to minimize the number of actions required to reach the coin in a short inference time!

Keywords: argparse , BFS, DFS, Dijkstra

maze_updater.py

This component receives the action.txt file and commits the action by update the bot in maze_metadata.json file. The command line to run maze_updater.py in terminal must follow this:

python maze_updater.py -i action.txt -o maze_metadata.json

Here, the -i is an argument which defines the input file, and -o defines the output file.

The loop of reading metadata, indicating action and updating metadata keeps the bot moving continuously in the maze until the bot get the coin. For instance, in the first step, with the given maze_metadata.json and action.txt, maze_metadata.json is updated like follow:

{
    "width": 10,
    "height": 10,
    "obstacles": [[0, 1], [1, 1], [2, 1]],
    "bot": [5, 7],
    "coin": [0, 0]
}

But the game does not end there, this function will also update the new position of the coin in the maze when the last coin disappear by generating the coin's position randomly. So continue with the example, when the bot reach [0;0] then coin could be:

{
    "width": 10,
    "height": 10,
    "obstacles": [[0, 1], [1, 1], [2, 1]],
    "bot": [0, 0],
    "coin": [7, 4]
}

And the game goes on and on.

maze_displayer.py

We can not just run our software without any validation. One of the best way is to develop a user-friendly UI for it. Design your own UI for maze runner and implement it withih maze_displayer.py. A good UI should have following:

Show the structure of the map.
Show the position of the coin and the bot.
Show the current step of the process.

For example, a simple UI could be like this:

Step 0
************
*o*        *
* *        *
* *        *
*          *
*          *
*        X *
*          *
*          *
*          *
*          *
************
Step 1
************
*o*        *
* *        *
* *        *
*          *
*          *
*       X  *
*          *
*          *
*          *
*          *
************
Step 2
************
*o*        *
* *        *
* *        *
*          *
*          *
*      X   *
*          *
*          *
*          *
*          *
************

Keywords: pygame, tkinter

We look forward to seeing not only your amzing results and a thrilling race but also your development through this very first project. Good luck!