wraggster
February 4th, 2008, 20:23
Heres a rather excellent release/news from the great DevFR forums and Ecaheti (http://www.dev-fr.org/projets/(ndsapp)pekeepilot/)
Hello everyone!
I helps me newser myself to share with you my Finally, our project of Completion.
For recituer context, as part of my engineering studies, I had a project to be carried out in a business or a personal project. The beginning of the project was February 2007. In December 2006 I discovered PAlib, and as a geek, I decided to do my project.
So I mounted with a vague pretext darling (student at my school too) to propose a project with the DS, namely drive a robot with the DS.
Stroke of luck, the school had a small robot shaped vacuum, Pekee, which wanted to use it to be out of his closet.
January, we recruited 4 persons (the project is done in teams of 6). From February to June 2007, we conducted studies on the various options and the achievement of tricks "innovative" to do with the console (with specifications, feasibility studies and the like).
From September 2007 to January 2008, we completed the technical part.
The basic idea:
6 steering interfaces (to send orders genre forward, backward, left, right, stop or otherwise)
-- Keyboard
-- Virtual buttons
-- Target (based on the handling of Mario in Mario 64 DS)
-- Spatial
-- Vocal
-- Recognition of form
Liaison between the DS wifi and Pekee.
In the end:
6 interfaces as planned, except that the voice recognition is replaced by a recognition "music", it plays a note to the flute, the note is recognized and corresponds to an order. The principle behind algorithmic is the same except that the calculation time is faster and easier identification.
The Liaison, which was to be reported directly to the intermediary of a Linux server. That's because the Pekee must be driven with its own library of programming that we have failed to focus on the DS. Broadly speaking, it runs a program that drives the robot.
Our demos were feeling at school, and now it's all over, let me offer expose the finished project.
At the project's future, I intend to continue to work primarily on voice recognition (improve algo.) Because the courses are finished, and I have more robot! The reco voice is the only thing that really needs to be improved.
The program can be used in offline mode for those who want to see what it looks like. I guess it's not much n'impressionnera world since it remains "a remote control."
And a small video because it's funny like that view
http://fr.youtube.com/watch?v=b_UHPoXOtfQ
(I tried to integrate video, not work, a nice if someone want to tell me how it's done!)
In terms of special thanks, of course Mollusk that we troubleshoot / help / support for a year. This guy sends wood.
PS: you have a site with some docs on the project, but it has not been updated since .. Pfiou .. Long. I shall update and I posterai link
Download and Give Feedback Via Comments
Hello everyone!
I helps me newser myself to share with you my Finally, our project of Completion.
For recituer context, as part of my engineering studies, I had a project to be carried out in a business or a personal project. The beginning of the project was February 2007. In December 2006 I discovered PAlib, and as a geek, I decided to do my project.
So I mounted with a vague pretext darling (student at my school too) to propose a project with the DS, namely drive a robot with the DS.
Stroke of luck, the school had a small robot shaped vacuum, Pekee, which wanted to use it to be out of his closet.
January, we recruited 4 persons (the project is done in teams of 6). From February to June 2007, we conducted studies on the various options and the achievement of tricks "innovative" to do with the console (with specifications, feasibility studies and the like).
From September 2007 to January 2008, we completed the technical part.
The basic idea:
6 steering interfaces (to send orders genre forward, backward, left, right, stop or otherwise)
-- Keyboard
-- Virtual buttons
-- Target (based on the handling of Mario in Mario 64 DS)
-- Spatial
-- Vocal
-- Recognition of form
Liaison between the DS wifi and Pekee.
In the end:
6 interfaces as planned, except that the voice recognition is replaced by a recognition "music", it plays a note to the flute, the note is recognized and corresponds to an order. The principle behind algorithmic is the same except that the calculation time is faster and easier identification.
The Liaison, which was to be reported directly to the intermediary of a Linux server. That's because the Pekee must be driven with its own library of programming that we have failed to focus on the DS. Broadly speaking, it runs a program that drives the robot.
Our demos were feeling at school, and now it's all over, let me offer expose the finished project.
At the project's future, I intend to continue to work primarily on voice recognition (improve algo.) Because the courses are finished, and I have more robot! The reco voice is the only thing that really needs to be improved.
The program can be used in offline mode for those who want to see what it looks like. I guess it's not much n'impressionnera world since it remains "a remote control."
And a small video because it's funny like that view
http://fr.youtube.com/watch?v=b_UHPoXOtfQ
(I tried to integrate video, not work, a nice if someone want to tell me how it's done!)
In terms of special thanks, of course Mollusk that we troubleshoot / help / support for a year. This guy sends wood.
PS: you have a site with some docs on the project, but it has not been updated since .. Pfiou .. Long. I shall update and I posterai link
Download and Give Feedback Via Comments