FEZ Mini Robot line maze solver

There are two runs, with the first run the maze and examined in the second run is the shortest path driven

This solution is the simple theory of Richard T. Vannoy see richardvannoy.info applied. This theory can only solve maze without looping.

Wow, that looks really great. What line sensors did you use?

Nice! Why did it pass up the one branch to the left? Was that a mistake or would it have come back to it if it hadn’t found the end?

I used as a line sensor CNY70 8 pieces in a row that mounted with two pieces connected to the PCF8591 I2C connector.
The sensors are supplied with well fitted but not used for the line maze.

Nice work!

Nice job!

This is fast! I have added it to the robot page if you do not mind.

I’m busy preparing a workshop on “Robot Programming with C #” and use the FEZ Mini robot to test the programs.
The program has been quite completed and exercises for the workshop, I have two movies that use the standard sensors:

Robots have been always fun to see. ;D

Great job Hinnie… If you place the light sensors 1 or 2 notchs closer, do you think it will make less wider turns?
This is a great project for your workshop… have fun out there.

Such great timing considering I just bought a FEZ Mini Robot yesterday :wink:

Are you going to post the source code for these?

@ twkisner [quote]Are you going to post the source code for these?[/quote]

I gave a link to the website of Richard T. Vannoy you the theory of the algorithm finds this help you much more than the code without much explanation.
My test code contains many features such as a link via comport with a PC so you can monitor what happens to the robot.
If you have problems, I want help with specific questions.

Bah Humbug :wink:

For what it’s worth, I’m going to try to port over the examples from the Inex POP-BOT manual and put them on the Wiki so there is a more complete guide to programming FEZ Mini Robot.

Hinnie,

Could you share your code ? I’m doing some classes to encourage young students to study Electronic Engineering and use Fez Mini Robot as a base.

I want to show them what could be done but dont have enough time to do it.

The problem of the “Line Maze” in the FEZ Mini Robot is not the algorithm. The algorithm used in this program is well documented theory of Richard T. Vannoy.
This I have already programmed in C for a ATmega32 that directly test work reliably without much work. Convert it to C # for the FEZ Mini showed testing initially unreliable work. Further analysis showed that by not “real time” of the FEZ prevent more problems than I had thought. For this I have several actions to be here less bothered by it.
As a solution for the random “active” from the Garbage Collection process in order to force it out of time-critical moments, so you can expect that this process does not start in a time critical part of the program.
At first I used to wheel rotation sensors to control the spin. The wheel rotation sensor works based interrupts, by counting the pulses is nothing wrong but it appeared that at the time of reading the “counter” it has not always been updated so you worked with a false value. Using the system clock gives a better result.

At our club we will soon launch a workshop on “Robot Programming with C #” and for that we use the FEZ Mini robot kit, there have been 20 participants signed up and have already bought FEZ Mini Robot.
The aforementioned movies are my preparation exercise results from the second workshop day. My idea was to linemaze exercise as homework after the workshop to give, but from my experience that seems too complex and no longer sensible.