-1

I am thinking about starting a new project with the purpose of mapping and navigating a maze with a cluster of robots. The number of robots I was thinking about are 2 or 3.

The following assumptions are made :

  1. The robots are fitted with a camera each to help detect the walls of the maze
  2. The size and shape of the maze is unknown and can be changed according to will
  3. The way the robots should work is that they should communicate and efficiently divide the task of mapping and navigation among themselves.

I am studying Electrical Engineering and have no previous experience with maze planning/solving with robotics. I would like to know as to how to begin with this; and more importantly the resources I should be looking at. Any suggestions of books, websites, forums are welcome.

The microcontroller I am planning to use is Arduino Uno. I am familiar with it and it has very good support online. Thus it seems to be a good choice. Also, I will have around 2 months to finish the project. Is that amount of time enough to accomplish the aforementioned things?

mysticsasuke
  • 119
  • 1
  • 4
  • 12
  • If you said 6 months, I would have been skeptical. But 2 months? And you don't even know which approach you want to take? I think 2 months would be difficult with only one robot. – ziggystar Nov 08 '14 at 13:11
  • I forgot to mention that I would be working with a team of 2 other people. It would probably be like 3 months and would it still be too much? – mysticsasuke Nov 08 '14 at 13:28
  • To begin with, do you have any idea how difficult it is to detect a wall with a camera? – ziggystar Nov 08 '14 at 13:46
  • No I do not. The simplest way to make the maze, and for the bot to recognise the maze is alright. Is the time too little? – mysticsasuke Nov 08 '14 at 14:01
  • How about asking your supervisor? It's hard to judge, not knowing what exactly you want to do, and what you already know. But by guess is yes - the time is wa too little. – ziggystar Nov 08 '14 at 14:07
  • I do have a teacher guide, though we have to make the proposal ourselves. Will ask the guide then. Thanks for the help though. – mysticsasuke Nov 08 '14 at 14:11

1 Answers1

0

A single robot in a maze is called a Braitenberg-vehicle. A group of such robots is a multi-robot-formation, which implies that the agents must coordinate their behavior. In the literature such games are called “Signaling games”, because a sender has private access to an event and must share this information with the group. For example, robot1 has detected a wall and sends the status update to another robot.

In it's easiest form, signaling games are modeled with a lexicon. That is a list of possible messages between the robots. For example: 0=detectwall, 1=walkahead, 2=stop. As a reaction to a received signal, the robot can adapt his behavior and change his map of the maze. Sometimes this idea is called a distributed map building algorithm, because the information is only partial available.

A basic example is the case if two robots are driving against each other (Chicken Game). They must communicate about their evasive strategy to prevent collision. If both robots have decided to drive in the same direction, they will collide as well. The problem is, that they don't know, what the other robot is planning.

Manuel Rodriguez
  • 734
  • 4
  • 18