Re: self-propagating RCX program
"Claude Baumann" <[email protected]>
| Newsgroups | gmane.comp.hardware.lego.robotics |
|---|---|
| Organization | Boarding School Convict Episcopal |
| Message-ID | <[email protected]> |
In lugnet.robotics, [email protected] wrote: > > On 09/09/05, Steve Hassenplug <[email protected]> wrote: >> On Fri, September 9, 2005 1:07 pm, danny staple wrote: >>> On 09/09/05, Steve Hassenplug <[email protected]> wrote: >>>> >>>> However, I think making a self-replicating robot from LEGO is not >>>> realistic. >>> >>> I think efforts to build a Lego Robot that builds simple Lego models >>> have been made, and sucessfully, but only for fairly simple assemblies >>> ( a 4 peice Lego "Duck"). >> >> You're right. If it's possible to design a robot that can build a four piece >> duck, >> it must be possible to design a robot that can build a robot that can build >> other >> robots. How many parts can these robots have anyway? Maybe a couple hundred. >> That's only fifty times as many as a duck. >> >> Still, I think I'll stand by my claim that making a self-replicating robot >> from LEGO >> is not realistic. >> >> Steve > > I agree with you that it is probably not realistic, but that wont stop > me having fun trying. > > I would like to build a robot that can build stuff, though I will > probably not use the RCX to control it - something with many more > outputs and sensors would be required. Still - we already build Lego > tools for stuff like pushing out pins and such - so although it is far > away, it is along a path people have already started down. > > I get the feeling though that building a robot to assemble using Lego > peices is easier than one to mould or assemble non-Lego. > > OrionRobots I'm more optimistic. This is extremely challenging. We should really start with an assemling robot. There are experiences to gain first. Hope there are some of the mechanical gurus that help think about it. What about you guys Ferrari? Or Philo? -Claude