Servos controlled by different controlller

[yash101]

I was wondering if some small servos, possibly used in a camera holder, could be controlled by an external computer like a propeller chip or raspberry pi. Is there a maximum wattage rating that can be controlled by an external chip.The computer will be doing vision processing so I think it will be more efficient for that to control the servos

[techhelpbb]

Originally Posted by team1165wins

I was wondering if some small servos, possibly used in a camera holder, could be controlled by an external computer like a propeller chip or raspberry pi. Is there a maximum wattage rating that can be controlled by an external chip.The computer will be doing vision processing so I think it will be more efficient for that to control the servos

From a technical standpoint of course.
From an FRC rules standpoint: I suspect yes and no.

You can absolutely use a Raspberry Pi, Parallax Propeller or Arduino to send information to the cRIO over the I2C, digital I/O (bit-bang SPI) or Ethernet ports (and possibly CAN/RS-232?). Therefore for you can comply with the FRC rules by instructing the cRIO to move the servos for you. Therefore all the field safety mechanisms are in place. This makes any direct servo interface kind of useless. So once you reach that realization you are down to sensors on this additional device and what it brings to the table.

A netbook or small laptop on the robot can be connected to USB cameras on the robot and do video processing and if one desired instruct the cRIO to move servos for the camera. The upside is that the processing of the video can therefore consume much less cRIO resources and you can use cheap USB cameras (handy in case someone destroys one).

The Parallax Propeller basically offers 8 'cogs' or cores to watch sensors. You can interface a CMOS camera chip to the propeller and use it for video processing. Course unlike a laptop you won't get handy things like OpenCV.

Keep in mind if you provide one of these additional items on the robot you need to provide it clean and reliable power. This has been an issue for some teams. A laptop generally comes with a battery and you are allowed to use it on the robot.

Team 11 has had the option of fielding a laptop for the last 2 years. We either found we performed well enough without potentially gaining 2 lbs or just didn't have time to risk too much live play testing. We did drive it over and over the obstacles in the field and so long as you do not have a mechanical hard drive it wears pretty well.

Aside to this:
I am curious where FIRST FRC stands on CAN/RS-232 interfaced systems. I know you are not allowed to control the Jaguars directly from something other than the cRIO but I personally have not seen anyone try to get inspected a robot that has a custom circuit on it that sends data over CAN to the cRIO and lets the cRIO echo back instructions to the Jaguars. I have also not seen anyone setup a multi-drop serial network which is entirely possible from an RS-232 port (see RS-422/RS-485).

[Chris_Elston]

Originally Posted by techhelpbb

Aside to this:
I am curious where FIRST FRC stands on CAN/RS-232 interfaced systems. I know you are not allowed to control the Jaguars directly from something other than the cRIO but I personally have not seen anyone try to get inspected a robot that has a custom circuit on it that sends data over CAN to the cRIO and lets the cRIO echo back instructions to the Jaguars. I have also not seen anyone setup a multi-drop serial network which is entirely possible from an RS-232 port (see RS-422/RS-485).

In 2012, we used a 6 axis IMU sensor that interfaced to the serial port of the CRIO. No problems inspecting.

Here is a picture of the interface box:
http://www.flickr.com/photos/teamthr...57629273847264

The sensor worked very well, but we kept blowing the voltage regulators on the circuit board. Because of this problem, we had to disable and by-pass the IMU sensor after it kept failing due to power problems.

[techhelpbb]

Originally Posted by Chris_Elston

In 2012, we used a 6 axis IMU sensor that interfaced to the serial port of the CRIO. No problems inspecting.

Here is a picture of the interface box:
http://www.flickr.com/photos/teamthr...57629273847264

The sensor worked very well, but we kept blowing the voltage regulators on the circuit board. Because of this problem, we had to disable and by-pass the IMU sensor after it kept failing due to power problems.

Very nice! Thanks for stepping forward.

I guess that leaves the question of whether anyone tried to pass a custom circuit interfaced by CAN to the cRIO through an official FRC inspection. Also whether or not they were using the Jaguar electronic motor controls over CAN on the same robot.