[Robotgroup] More USB info

Sat Jun 2 23:46:39 PDT 2007

I did buy a couple of WRT54G (pre v5) with the intention to hack
them eventually. I don't have any of the WRTSL54GS but they also do
look like a good match for robotics.

I don't see a problem with the id changing if you implement 
serialization (as in serial numbers) of your hardware. The
difficulty under such scenario would be to figure out which
arm the hand is attached to, specially since all devices would
connect to the central brain (ie. the SBC with USB host controller).
If you have a simple arm id that the hand would know what it is
when connecting to it, then it can report to your driver when it asks
for its id then the problem is solved.

It does propose an interesting concept in which you could plug
20 hands into a single arm. Quite an intriguing control problem.

I never got my USB 802.11 cards working under Linux so they've
just been collecting dust. It seems that you have an interesting
use for them.

Alberto

On Sun, 2007-06-03 at 01:23 -0500, Glenn Currie wrote:
> Well I  hope the USB information is useful.  There is a lot of USB 
> equipment out there
> for very reasonable prices.
> 
> For some ham radio related projects, that are directly usable with some 
> robot
> projects, we have had great results with the Linksys WRT54GL.  Linksys
> started out using Linux to get a a free TCP/IP stack.  They also made use
> of the host USB drivers in some of their projects like the WRTSL54GS.
> 
> The WRTSL54GS only had one USB port, but it does support a hub,
> if you plug one in.  I have a D-Link 7 port powered hub (has its own
> power supply so it can deliver full power to 7 USB ports.)  It works
> great.  I plugged in a bunch of USB disk drives, and thumb drives
> and I could mount each file system.  I was also able to use the device
> to make an interesting 802.11b/g sniffer.  I hooked 5 USB 802.11b/g
> adapters into the hub.  Each radio was mounted in a hacked microwave
> horn, thus making each radio directional.  It was trivial to run 5
> sessions of NetStumbler on a laptop.  With 5 directional radios
> ( basically in a front, rear, left, right and strait up arrangement)
> and 5 data streams of signal strength and GPS coordinates, I can
> collect essentially 3-D (5-D?) data on access point locations.
> 
> I still have to correlate the data sets, that is some work but relatively
> straight forward.  Fry's runs specials on the Airlink101 model ASLL3026,
> USB to 802.11b/g adapters for $9.95.  These are fantastic little radios and
> there is a lot of interesting hacking that can be done with them.
> 
> The mass produced embedded Linux devices could be very useful in
> robot projects.  They have serial ports, some have two, and there is
> JTAG support as well (you have to solder in headers for these ports,
> but the support for them is built into every box.)  There is information
> on the net about doubling the RAM space, as well as many other mods.
> Most of the hackers working on these devices are network folks, but
> I have seen several projects where they were used for robotic devices.
> 
> I think it's great that you are looking for ways to use USB devices.  I 
> learned
> a lot about USB several years ago while writing some drivers.
> It turned out to be a lot more work than
> I had expected.  If it had not been for the $14,000.00 USB bus analyzer we
> had, I would have never gotten all the code to run.
> 
> Jan Axelson's books are quite good.  I find it amusing / telling, that his
> "USB Complete" book is in it's second edition - I guess it wasn't so
> complet now was it? ;-)
> 
> Actuallly the second edition coveres the many changes for the USB 2.0 spec.
> 
> USB was originally not intended for high bandwith connections.  One of
> it's original goals was to be inexpensive.  By providing a reasonable amount
> of power on the bus, peripherals would not need a power supply.
> 
> There were some business / people issues about the use of firewire, so some
> of the key players pushed for a faster USB spec. so they could compete.
> 
> When I first started working with USB devices, the specs looked impressive.
> I had been trying to work out a modular, self configuring peripheral bus for
> robots.  The idea was that things like arms, hands, sensors, including 
> video,
> could be hot plugged and the master controller could automatically configure
> the robot to make use of the new device / limb.  USB seemed like a good
> fit for a while.  It supplied enough power to run a microprocessor in each
> device, and I had planned for a separate power bus for motors and other
> actuators anyway.  (Then there was the mechanical coupling to work out.  
> I wanted
> some sort of quick connect disconnect - another project / tangent.)
> 
> Once I started working with USB it became clear that, in theory, it was
> a good fit for what I wanted, but the implementation on the OS's at the
> time was pathetic.  Things have gotten much better, but I still have 
> problems
> with my USB to RS-232 adapters.  They enumerate OK, but depending on
> what else I have plugged into the computer, it can show up as a different
> numbered serial port - most of them above COM4.  A lot of perfectly good
> software can only address comm ports 1 - 4, so when my adapters
> enumerate to COM7, such software is useless.  With enough work,
> I can usually get a system configured so that the USB serial port is usable,
> but then it is not really "hot pluggable" and self configuring is it?  
> It is just a
> new and complex pain the the ass.
> 
> Perhaps some  industrial strength variation to USB would be workable, but
> as it is now, if I plugged in an "arm" module it might auto configure to
> a leg, an ear or a finger.  A hand on an arm is useful, a finger in the ear
> is only useful, when you need ear plugs. ;-)
> 
> Somehow it brings to mind a situation where  Bill Gates in in the
> hospital - and the life support system attached to him,
> is running Windows. 
> 
> Can you imagine the horror he would be fill with as the
> "Blue Screen of Death" replaced his EKG pattern on the monitor?
> 
> Cheers,
> 
> 
> -Glenn
> 
> 
> >
> >   
> 
> 
> _______________________________________________
> Robotgroup mailing list
> Robotgroup at puremagic.com
> http://lists.puremagic.com/cgi-bin/mailman/listinfo/robotgroup
-- 
Alberto Alonso                        Global Gate Systems LLC.
(512) 351-7233                        http://www.ggsys.net
Hardware, consulting, sysadmin, monitoring and remote backups