Control System Emulator

[Madison]

I am asking this question on behalf of some of my team's programming mentors and students, so please keep in mind that I don't really know what I'm talking about when it comes to these things and I might not communicate things clearly the first, third or eighteenth time.

We're in search of an emulator for the latest major iteration of the robot controller that will allow our programming team to develop and test code in the absence of a robot. Specifically, they've asked for something that does a good job of mimicking the robot controller's limitations in memory and computing power. They don't want to be able to simulate more than the real processor is capable of doing.

I recall from years ago that Rob Bayer had produced something like this and did a search of the forums this morning for information about it, but the links to his website (www.robbayer.com) and the associated emulator are dead. Does anyone have the last revision of this lying about somewhere? I believe it was updated to reflect the change to the C programming language.

Thanks!

[Jon236]

Madison,

I believe LabView is capable of doing what you are asking. You can watch the ouputs in terms of defined motor and/or actuator devices and use real or simulated inputs. Check out these threads

http://www.chiefdelphi.com/forums/fo...play.php?f=149

[Madison]

Jon,

Thanks I'll pass this along to the programming folks and see if this will meet their needs. My impression was that they wanted something that could be used completely independent of a robot controller -- so students could develop and test code from their homes -- but maybe this will be a viable alternative.

[dtengineering]

I have found that my students who use the VEX controller (yeah, it's hardware, not exactly the solution you're looking for) find migrating their code to the RC is pretty easy.

On the other hand, something suitable may be coming for the future.

I say that with some hesitancy, due to this thread, and do not want to drag this off track into a discussion of what next year's RC might include. It merely seems -- based on the description of MS Robotics Studio -- like the technology would be capable of supporting excellent software prototyping if the FRC components and control system were appropriately modelled in it.

Jason

[chris31]

I dont know if you can yet, since I am no LabVIEW pro, but I think NI's goal was to allow you to fully test these types of things without any physical robot in front of you.

[Phalanx]

Originally Posted by M. Krass

I recall from years ago that Rob Bayer had produced something like this and did a search of the forums this morning for information about it, but the links to his website (www.robbayer.com) and the associated emulator are dead. Does anyone have the last revision of this lying about somewhere? I believe it was updated to reflect the change to the C programming language.
Thanks!

I have a copy of what I believe to be the last known version of it.
It's RoboEMU2-0.1
I've uploaded it for those that may desire to look at it.
I've never tried it or used it. Use at your own risk.

Any feedback of those that explore it would be of great benefit.

[Nathan]

I have a Vex kit. The controller is almost identical to the FRC one, it can even use FRC sensors and the CMU cam.

[Kingofl337]

I also agree that VEX is the best emulator you could ask for. If only the controller could run off serial or USB power.

[Qbranch]

Or you could just use the MPLAB SIM thats build in to the development environment... just define "simulator" in the code... you'll see where IFI has made provisions for it... after that you can build stimulus tables and simulate inputs.

-q

[3dude_2231]

Qbranch,
that is indeed what I used to do when I didn't have bot access.
but.. it is a frustrating procedure, watching the variables window,
going through simple delay loops can be very exhausting,
then, remembering that you forgot something in the stimulus section,
and going through the whole thing again

I now only use it in small application (with my PIC16s), and it's hard enough =]

I have heard of MS's "robotic plans", but I didn't know something like the roboEMU exists, I'll be checking this one right away =]

[ham90mack]

See http://www.chiefdelphi.com/forums/sh...ad.php?t=59039

[Chief Pride]

Originally Posted by Kingofl337

I also agree that VEX is the best emulator you could ask for. If only the controller could run off serial or USB power.

i am thinking of developing something like that, a quick 3d "game" using the robots actual joysticks and controllers, drivers could gain alot more experiance with the way the controls work

however, nothing will beat the real thing

[Qbranch]

Originally Posted by zinefer

i am thinking of developing something like that, a quick 3d "game" using the robots actual joysticks and controllers, drivers could gain alot more experiance with the way the controls work
however, nothing will beat the real thing

Well, one thing you can try soon is the real time physical simulator in SolidWorks cosmosMOTION... its currently only available in the non-educational version but very soon it will be available for all. The new API based interface to cosmos lets you write code for a machine and see it run as it really would in a full physical simulation... you can do everything from size motors to figure gear ratios in full physical simulation before your robot ever leaves the modeling space.

-q