Subject:
|
Re: Microsoft Robotics Studio
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Sat, 24 Jun 2006 15:49:20 GMT
|
Original-From:
|
PeterBalch <PETERBALCH@COMPUSERVE.COMstopspammers>
|
Viewed:
|
2351 times
|
| |
|
|
I guess my postings (here and elsewhere) have been amongst the dozen that
are way off topic. But I still don't get it.
What is MSRS giving me that I want?
It may be well thought-out and implemented but so what? I'll certainly look
closely at the architecture they've chosen so as to to pick up tips. But
I'm very unlikely to use their system.
I don't want to build a robot whose brain is so enormous it can run XP
.Net. And I don't want a robot that's a dumb terminal.
So let's say I've got an autonomous robot that's not got a huge barin. What
will MSRS do for me? It can show debugging information on the PC screen so
long as I've written some really rather complicated code on my robot and
given it Bluetooth or somesuch.
Thanks but no thanks. I've never found it difficult to display debugging
info on my PC. That sort of program is always trivial to write. And it's
always trivial to write the sort of comms I want to run on my robot.
Typically, in half an hour or so, I can get a bare PIC sending data to my
PC via radio or IR or whatever and the PC drawing graphs. There's no way I
can implement Bluetooth or 802.11 on a PIC in half an hour.
So what's left in MSRS? A simulator.
It sounds like it's a good, free mechanical simulator that can also run
small C programs. That's great.
But only if it really is a GOOD mechanical simulator. Can it simulate, say,
the mechanics of a hopping robot that occasionally bumps into the furniture
of slips on the polished floor? It sounds like it can. If it has that level
of sophistication, then it will be a great toy to play with.
But for robot design? I don't find that simulation helps all that much. I
understand why large companies use CAD and simulation when designing
products: whole airliners nowadays are put into production without a
physical prototype. But I've never found that level of simulation useful.
If I want to fit a motor, some cells, a pcb and a gearbox into a space, I
find it much easier just to fiddle around with the actual objects. CAD gets
in the way. And simulation? I'd probably use it for walking but not for
anything else. Can it simulate false sonar echos? Can it simulate vision
sensors being blinded by a specular reflection? Simulation is great for
designing a rocket to Mars but not for going to the corner shop.
What about Education? My first thought was that the simulator would be
great but, on reflection, no. It would be yet another nail in the coffin of
reality-based education. I don't know how things are in other countries but
here in the UK, a child can go right through school learning physics,
chemistry and biology and never perform an actual experiment. The closest
they get is to watch an interactive animation of an experiment. For me, the
outstanding attraction of robotics is that it's physical. The whole world
runs only because mechanical engineers make mechanical things happen.
What's going to happen in 20 years when children whou've never poked
themselves in the hand with a screwdriver are trying to build and repair
power stations? The UK politicians are assuming we'll just import real
engineers from Eastern Europe.
So I'd ban the simulator from education.
That's off-topic rant number 13.
Peter
|
|
Message has 2 Replies:
 | | Re: Microsoft Robotics Studio
|
| (...) Another potentially useful environment and different way to design a robotics program. You may not like it, you may love it! SDK's and IDEs are somewhat religious topics among developers. (...) Huh, you now if I didn't know better, I would say (...) (18 years ago, 26-Jun-06, to lugnet.robotics)
| | | Anyone had success with Microsoft Robotics Studio
|
| Hi everyone Is there anyone on this list who has managed to get the canned demo of Microsoft Robotic Studio working with NXT? Sounds simple, but several of us with an interest in this have tried. No one has been able to get it to stay connected for (...) (17 years ago, 31-Jan-07, to lugnet.robotics)
|
5 Messages in This Thread:
  
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
