I'll soon start working at a maze solving robot. And right now I'm trying as much as possible to search the proper algorithm that will stand at the base of my future code. So I'm asking you guys for some papers or something related that will help me. :smile:
Let me be more specific now. I want the robot to first evaluate the maze (with walls) by consistently moving around (turns left & right, back and forward) until it reaches the end. But I want it to be able to remember the shortest route which can be recalled at a later date. Did anyone read a practical implementation (paper) of such a robot ?
You could try A* or B*. you could get more info from Wiki about A* search algorithm. For collision avoidance along the path you could use Vector field Histogram(VFH). We tried A* with VFH some time back and it worked well.
**broken link removed**
A walk through of the line maze algorithm.
**broken link removed**
A PowerPoint slide show, in PDF format that takes you through the line maze algorithm in great detail. I make this a "Must read" for my students building line maze robots.