We got some good feedback on the Mascot Robot at the december meeting, and there was an impromptu "Mascot design / BS" meeting afterwards at the nearby Dairy Queen.
Comments:
Provide Modularity - This is an effort to make the bot easier to maintain and modify in the future. With different people working on different parts, parallel tracks can be taken, different configurations explored simultaniously, etc. The controller that First uses was discussed because it provides Pendant, MCU and RF control . Does anyone have access to one of these from a retired First Bot? Something to think about is how to make not only the electronics modular, but to standardize on a way to make the mechanicals modular as well. Initially I will just proceed with simple triangle and rectangle modular sections that can be joined at the rotation axis' where the articulations join. (does that make any sense?)
ATmega - Mike Panetta offered two ATmega boards to the project.
Provide space for optical encoder - John Scott has offered to provide "optical mouse" encoders for the bot so space must be provided in the design. I'll put that into the CAD file once results of Dale's modifications come in.
Power transmission - Dale Heatherington took the gearbox and wheel home to try out some different coupling ideas.
Discussion - Discussion spun outward from there into subjects like laser obstacle avoidance ( Home Depot has a new laser range finder $400 ), Ultrasonic local GPS "LPS" schemes (this one is a ripe fruit and not too many others have attempted it), and many other subjects.
A couple of ideas I found particulary intresting. One was to investigate adding a very high level programming interface like Logo ( FlowCode was reccomended by Mike Lynch) so that newbie club members could jump in and get programming quickly on the mascot. Another idea was the posibility of making the mascot "balancing". There are more and more examples of this on the web and in other clubs, and proven code is available. The cost of the required gyroscopes and accelerometers seem to be the main barrier, but this is bound to be a parallel track in the future as those components become available.
3 comments:
As far as steering goes, differential looks like it
will provide the most movement flexability and easiest
movement control.
Using the front wheels of the 'Barbie' car sounds like
we have to develop control code to perform stearing
like a auto. That means if we get into a hallway and
need to turn around we have to perform something like
a 3-point turn.
Or maybe it would look more like a 3943-point turn ;)
-john
There is a good tutorial on robot recharging here:
http://www.psiaz.com/schur/astro/robotics/robotchargers1.html
I know Ringo has been working on this lately. Maybe he has some input in this area...
Keith
In reference to the modular design, there is a generic layout for such a system here: http://linuxdevices.com/files/misc/oap_block_diagram.png
Post a Comment