Johnny 5 Alive?!?
My nabaztag hasn’t been alive for quite some time, mainly due to the fact that it got damaged in my last house move about a year ago, but today I finally got round to repairing it. Checking out the current state of the nabaztag community got me thinking - time for me to re-wire my rabbit.
For those of you who don’t know what the nabaztag is, it was a little wifi enabled robot with two little moving ears that make it look like a bit like a rabbit. Current method of interaction with the rabbit currently consist of a button on the top of it’s head, 4 separate lights that can display multiple colors, and RFID reader, speaker, and microphone.
Normally the rabbit receives instructions on what to do via signals via wifi from the nabaztag website. The current problem seems to be that the nabaztag crew have never really delivered on what they promised. Exciting new features promised years ago are still missing, partially implemented, or full of errors. The fact that the company’s own development blog has been deleted, and that most of the fan sites in the community are ether closed or haven’t seen an update for a very long time seems to suggest that the future for these little robot rabbits is not looking good.
While I’ve managed to repair my rabbit enough to re-connect to the service website and thus receive instructions, currently the rabbit’s ears do not work correctly (they’re supposed to move as well. This means that this little device now falls into what I see as the prime hacker spot. The area where a gadget still has some use, but is for the most part now pointless with no foreseeable future unless you upgrade it.
From what I know about the rabbit from past research, its not really something I could simple install linux on and have an uber-nabaztag, so I’m planning for something a bit more extreme. Soldering iron, and new components will be needed. I’m essentially looking at turning the rabbit into a mini-PC with a lot more processing power, memory, and a custom linux kernel. It will be a long project and if you know of anyone that had actually got round to doing this (and not just talked about it) please let me know via email.
If I get to do what I want - then the software should be obvious. Real internal voice recognition and speak synthesis. Internal AI that responds to data on the Net, instead of just being fed instructions from the. This should be fun! :)