View Full Version : Is the suggested Vision Camera setup ILLEGAL?!
The suggested camera configuration available from: http://www.ifirobotics.com/docs/cmucam2-documents-and-software1.zip requires a servo that pans the camera left and right to be directly connected to the camera board. (see GIF attachment)
Is this illegal according to <R53>?
<R53> Custom Circuits may not:
• Interfere with the operation of other robots
• Directly affect any output devices on the robot, such as by providing power directly to a motor, supplying a PWM signal to a speed controller or supplying a control signal to a relay module. (Custom high impedance voltage monitoring or low impedance current monitoring circuitry connected to the robot’s electrical system is acceptable, because the effect on the robot outputs should be inconsequential.)
• Be used for wireless communication, such as sending or receiving a signal to and/or from the alliance station
• Connect to the radio or tether ports on the Robot Controller
-SlimBoJones...
Kris Verdeyen
14-01-2005, 12:07
An interesting question - Especially when you consider that the camera itself might be able to cause the servo to move even if the robot is disabled. The servo motor will have power, and I'm pretty confident that a sufficiently clever FIRSTer could hack the camera board (in a legal way, of course) to cause the servo to move (!) when the robot should be disabled.
Has anyone tried this to see if it's possible?
An interesting question - Especially when you consider that the camera itself might be able to cause the servo to move even if the robot is disabled. The servo motor will have power, and I'm pretty confident that a sufficiently clever FIRSTer could hack the camera board (in a legal way, of course) to cause the servo to move (!) when the robot should be disabled.
Has anyone tried this to see if it's possible?
It is possible, and it DOES move when the robot is disabled. No hacking is necessary, as this behaviour was observed using the supplied Frc_CMUCam2_rev-3.zip code package on the IFI website.
To recreate:
1) Disable OI with competition port dongle
2) Power on robot
3) Select desired colour by pressing Trigger on Joystick 2
Watch camera servo move and track your colour while the robot is disabled.
Interestingly enough, this turned out to be "a feature, not a bug" during camera setup. It was nice knowing your robot wasn't going to bolt on you, yet you could see if your camera was calibrated/configured properly by watching it move.
-SlimBoJones...
Matt Leese
14-01-2005, 12:53
Except the camera isn't a custom circuit -- it's provided in the kit of parts. Hence it should be legal.
Matt
Katie Reynolds
14-01-2005, 13:15
Except the camera isn't a custom circuit -- it's provided in the kit of parts. Hence it should be legal.
MattYep.
Follow the flowchart (Sec. 5, Pg. 14):
Is the part a safety hazard or likely to damage robots, the field or interfere with homes or the controls?
|
NO
|
Bumper or non-functional decoration?
|
NO
|
Kit Part?
|
YES
|
Yes, the part may be used.
Yep.
Follow the flowchart (Sec. 5, Pg. 14):
Is the part a safety hazard or likely to damage robots, the field or interfere with homes or the controls?
|
NO
|
Bumper or non-functional decoration?
|
NO
|
Kit Part?
|
YES
|
Yes, the part may be used.
It's not a question of whether the part itself can be used, but whether it is allowed to be connected in the manner described in the document above.
-SlimBoJones...
Gal Longin
14-01-2005, 13:34
I would say according to the rules it's legal but it's useless.
in order to track the vision tetras, your robot should be able to move and when the robot is disabled it can't move.
another way which might work would be to connect the servo outputs from the camera to the analog inputs in the RC and then control the servo through the PWM output, in this case u can also use the signal from the servo to monitor the heading of your camera , which can be quite useful using some "smart" programming.
I would say according to the rules it's legal but it's useless.
in order to track the vision tetras, your robot should be able to move and when the robot is disabled it can't move.
another way which might work would be to connect the servo outputs from the camera to the analog inputs in the RC and then control the servo through the PWM output, in this case u can also use the signal from the servo to monitor the heading of your camera , which can be quite useful using some "smart" programming.
Gal,
I would disagree. It would be useful to know that your robot is successfully tracking the tetra right before the match begins (while it is disabled). In the event that it isn't, you can quickly switch to an autonomous mode that doesn't use the camera, instead of wasting 15 precious seconds.
Additionally, the camera would get its lock BEFORE the match starts instead of having to waste time doing it after it has already started.
Lastly, this is the method IFI is SUGGESTING in the "Start Here" document. This means most teams setting up these cameras for the first time, will be setting them up this way. It'd be errr... interesting... if this way turns out to be illegal.
-SlimBoJones...
Dave Flowerday
14-01-2005, 14:45
Hopefully we'll have a definitive answer to this soon:
ID: 1183 Section: 5.3 Status: Unanswered Date Posted: 1/14/2005
Q: May we connect servos to the PWM outputs on the CMUcam to take advantage of the automatic pan and tilt control that it provides?
Tom Bottiglieri
14-01-2005, 14:49
Wow, this could really be useful.
This could cut autonmous times by 1-2 seconds.
Good Find.
seanwitte
14-01-2005, 14:56
The cam data structure in the sample camera code includes the values for the pan and tilt servo commands when in tracking mode. Instead of plugging the servos straight into the camera board you can grab them from the cam struct and use the PWM outputs on the RC to control the servos. Its not really fair to scanning the field before the match starts is it?
Gal Longin
14-01-2005, 14:56
Gal,
I would disagree. It would be useful to know that your robot is successfully tracking the tetra right before the match begins (while it is disabled). In the event that it isn't, you can quickly switch to an autonomous mode that doesn't use the camera, instead of wasting 15 precious seconds.
Additionally, the camera would get its lock BEFORE the match starts instead of having to waste time doing it after it has already started.
Lastly, this is the method IFI is SUGGESTING in the "Start Here" document. This means most teams setting up these cameras for the first time, will be setting them up this way. It'd be errr... interesting... if this way turns out to be illegal.
-SlimBoJones...
wow, i have to say i didn't think of this in that way. you're absoloutly right this could really save you a lot of time during the autonomous mode. thanks a lot.
Gal,
I would disagree. It would be useful to know that your robot is successfully tracking the tetra right before the match begins (while it is disabled). In the event that it isn't, you can quickly switch to an autonomous mode that doesn't use the camera, instead of wasting 15 precious seconds.
-SlimBoJones...
Except for the fact that you cannot touch your controls and robot after the vision tetras have been placed per rules
<G05> The VISION TETRAS will be placed on the field by field attendants after the ROBOTS have been
positioned. The locations of the VISION TETRAS will be determined by random selection from a set of 8
possible starting locations (as identified on the “2005 Field Lines and Layout” drawing). ROBOTS may not
be repositioned or manipulated in any manner after the VISION TETRAS have been placed on the field.
<G06> No team member may pass the Starting Line in their TEAM ZONE until the conclusion of the
AUTONOMOUS PERIOD. All team members must stay within their alliance’s designated TEAM ZONE
during the match. If a team member passes the Starting Line before the autonomous period ends, except to
save their controls from a violent collision of a robot into the diamond plate, or leaves their TEAM ZONE
during the match, the team will be assessed a 10 point penalty. If a HUMAN PLAYER leaves the HUMAN
PLAYER ZONE at any time during the match for any reason other than personal safety, the team will be
assessed a 10 point penalty. However, if such actions are deemed by the referee to be so serious to have
affected the outcome of the match, the team may be disabled and disqualified.
Dave Flowerday
14-01-2005, 17:40
Its not really fair to scanning the field before the match starts is it?
As long as FIRST makes it clear that it's allowed (if that's their ruling, of course) then I don't see how it wouldn't be fair... every team would then have the opportunity of doing it, right? It seems like that would make the game better for everyone. Wouldn't it be more exciting if the robots knew exactly where to go as soon as autonomous starts rather than having to scan for them after getting enabled?
Dave Flowerday
14-01-2005, 17:42
ID: 1183 Section: 5.3 Status: Answered Date Answered: 1/14/2005
Q: May we connect servos to the PWM outputs on the CMUcam to take advantage of the automatic pan and tilt control that it provides
A: Yes.
We have our answer.
seanwitte
14-01-2005, 17:42
Except for the fact that you cannot touch your controls and robot after the vision tetras have been placed per rules
<G05> The VISION TETRAS will be placed on the field by field attendants after the ROBOTS have been
positioned. The locations of the VISION TETRAS will be determined by random selection from a set of 8
possible starting locations (as identified on the “2005 Field Lines and Layout” drawing). ROBOTS may not
be repositioned or manipulated in any manner after the VISION TETRAS have been placed on the field.
<G06> No team member may pass the Starting Line in their TEAM ZONE until the conclusion of the
AUTONOMOUS PERIOD. All team members must stay within their alliance’s designated TEAM ZONE
during the match. If a team member passes the Starting Line before the autonomous period ends, except to
save their controls from a violent collision of a robot into the diamond plate, or leaves their TEAM ZONE
during the match, the team will be assessed a 10 point penalty. If a HUMAN PLAYER leaves the HUMAN
PLAYER ZONE at any time during the match for any reason other than personal safety, the team will be
assessed a 10 point penalty. However, if such actions are deemed by the referee to be so serious to have
affected the outcome of the match, the team may be disabled and disqualified.
Whats hes saying is that with the servos attached to the camera module it will track the current color by itself. If the servos are not connected to the RC they they will not be disabled before the start of the match. They could potentially lock and track the tetras as they're carried out on the field. Whether it would work in practice is another issue....
This question has been answered via Official Q&A.
The suggested Vision Camera setup is LEGAL.
Moderators may close this thread.
ID: 1183 Section: 5.3 Status: Answered Date Answered: 1/14/2005
Q: May we connect servos to the PWM outputs on the CMUcam to take advantage of the automatic pan and tilt control that it provides
A: Yes.
We have our answer.This answer seems to be contradicted by the following answer.
ID:1184 Section:5.3.5 Status:Answered Date Answered:1/14/2005
Q:The suggested camera configuration available from: www.ifirobotics.com (http://www.ifirobotics.com/) requires a servo that pans the camera left and right to be directly connected to the camera board. Is this illegal according to <R53?
A:Yes.
Thoughts?
Oh great. What do I do now?
Dave Flowerday
16-01-2005, 17:10
This answer seems to be contradicted by the following answer.
Yeah, I saw that too... I'm hoping they read the second one wrong - perhaps they thought it said "legal" instead of the actual word "illegal". I can't imagine how they could intentionally answer those questions differently, seeing as they're consecutive question numbers AND they were answered within hours (maybe even minutes) of each other.
I guess someone will need to put up yet another question, saying that the answers to 1183 and 1184 contradict each other and could they please clarify. I hate to have to ask them redundant questions (because I know they're busy enough), but I'm not sure what else to do.
Greg Marra
16-01-2005, 20:04
if (cameralock == 0) {
deadreckon = 1
}
if (cameralock == 0) {
deadreckon = 1
}
programmerStress++; //:p
Joe Ross
17-01-2005, 15:16
I guess someone will need to put up yet another question, saying that the answers to 1183 and 1184 contradict each other and could they please clarify. I hate to have to ask them redundant questions (because I know they're busy enough), but I'm not sure what else to do.
The answer to 1184 has been removed. I guess that means the answer to 1183 stands and it is legal.
Wait, the camera should not be illegal, servo's are not relays, motors, or motor speed controllers. Running the servos through the IFI controller effectivly makes the camera a waste because of the slow update time.
As it points out in the camera module, the servos should be connected directly to the camera.
Wait, the camera should not be illegal, servo's are not relays, motors, or motor speed controllers. Running the servos through the IFI controller effectivly makes the camera a waste because of the slow update time.
Servos most definitely are motors. The fact that the angular rotation of the output shaft can be precisely controlled does nothing to change its status as a motor.
That aside, I look at the use of the servos as documented in the IFI documentation as components within a single assembly of kit components that FIRST has told us to build. I think that thier use in this context would be OK. Let's see what FIRST says.
-dave
We have our answer (again):
ID: 1184 Section: 5.3.5 Status: Answered Date Answered: 1/17/2005
Q: The suggested camera configuration available from: www.ifirobotics.com requires a servo that pans the camera left and right to be directly connected to the camera board. Is this illegal according to <R53?
A: No. (Answer CHANGED 1/17/2005) Servos connected to the camera module are considered part of the camera module and can only be connected to it.
RIgnazio
20-01-2005, 13:41
But do not forget this:
You CAN NOT use the camera servo outputs to power the angling servos for the camera. If you wish to have the camera move on its own, those outputs must go through the FRC Controller, and the FRC Controller must control the servos.
See Section 5: The Robot (http://www2.usfirst.org/2005comp/Se...5-The_Robot.pdf)
<R53> Custom Circuits may not:
• Interfere with the operation of other robots
• Directly affect any output devices on the robot, such as by providing power directly to a motor, supplying a PWM signal to a speed controller or supplying a control signal to a relay module. (Custom high impedance voltage monitoring or low impedance current monitoring circuitry connected to the robot’s electrical system is acceptable, because the effect on the robot outputs should be inconsequential.)
• Be used for wireless communication, such as sending or receiving a signal to and/or from the alliance station
• Connect to the radio or tether ports on the Robot Controller
Just to help you avoid headaches in the future.
Also, If you go to www.ifirobotics.com and look at the details for the camera, it states for the servo outputs "Not typically used in FIRST competitions" or something to that nature.
Rocketboy
20-01-2005, 13:45
Thanks for the heads-up :)
Alex1072
20-01-2005, 13:48
This may be a technicallity, but I thought if it's included in the kit, then it's not a custom circuit? Since all teams get one of these in the Kit, noones at a disadvantage. Is there something I'm missing?
RIgnazio
20-01-2005, 13:49
No problem. I had posted in a section with Programming, but the topic was closed. (I guess I was a little off topic) I was told to post in Rules.
Dave Flowerday
20-01-2005, 13:52
You CAN NOT use the camera servo outputs to power the angling servos for the camera.
Yes you can. There have been several threads on this subject already, and it has been asked and answered on the FIRST Q&A.
{edit} Looks like someone merged this with the other thread that was already talking about this subject after I posted this. {/edit}
ID: 1183 Section: 5.3 Status: Answered Date Answered: 1/17/2005
Q: May we connect servos to the PWM outputs on the CMUcam to take advantage of the automatic pan and tilt control that it provides
A: Yes. Servos connected to the camera module are considered part of the camera module and can only be connected to it.
I have merged these 2 threads to make 1 of the same. :)
RIgnazio
20-01-2005, 16:44
Pretty much, FIRST has made it confusing, once again. according to <R53>:
<R53> Custom Circuits may not:
• Interfere with the operation of other robots
• Directly affect any output devices on the robot, such as by providing power directly to a motor, supplying a PWM signal to a speed controller or supplying a control signal to a relay module. (Custom high impedance voltage monitoring or low impedance current monitoring circuitry connected to the robot’s electrical system is acceptable, because the effect on the robot outputs should be inconsequential.)
• Be used for wireless communication, such as sending or receiving a signal to and/or from the alliance station
• Connect to the radio or tether ports on the Robot Controller
it states that it can not control an output ON THE ROBOT.
Obviously, the camera is going to be on the robot, and not hovering or anything. (If you can get your cam to hover, let me know how! :yikes: j/p)
Now, according to the Q&AID: 1183 Section: 5.3 Status: Answered Date Answered: 1/17/2005
Q: May we connect servos to the PWM outputs on the CMUcam to take advantage of the automatic pan and tilt control that it provides
A: Yes. Servos connected to the camera module are considered part of the camera module and can only be connected to it. it is quite alright. I think someone needs to e-mail FIRST. I personally believe the rulebook, and am surprised that it hasnt been included in the latest FIRST rules update.
Pretty much, FIRST has made it confusing, once again. according to <R53>:
it states that it can not control an output ON THE ROBOT.
Obviously, the camera is going to be on the robot, and not hovering or anything. (If you can get your cam to hover, let me know how! :yikes: j/p)
Now, according to the Q&A it is quite alright. I think someone needs to e-mail FIRST. I personally believe the rulebook, and am surprised that it hasnt been included in the latest FIRST rules update.
Actually, it seems quite clear. Rule <R53> has to do with custom circuits. Question 1183 makes it clear that FIRST considers the CMUcam II system - including the connected servos - to be all one module that is supplied in the kit. It is not a "custom circuit." Therefore, <R53> does not apply. Thus, connecting the servos to the camera, as detailed in the supplied documentation from FIRST, is legal, and <R53> and Q 1183 are perfectly consistent.
-dave
Rickertsen2
20-01-2005, 18:31
It is not a "custom circuit." Therefore, <R53> does not apply. Thus, connecting the servos to the camera, as detailed in the supplied documentation from FIRST, is legal, and <R53> and Q 1183 are perfectly consistent.
-dave
Ah but by that reasoning what prevents us from connecting the speed controllers to digital IO and using them to generate a PWM signal?
Dave Flowerday
20-01-2005, 20:04
Ah but by that reasoning what prevents us from connecting the speed controllers to digital IO and using them to generate a PWM signal?
Probably the part of their answer to the Q&A that says "Servos connected to the camera module are considered part of the camera module and can only be connected to it." Let me rephrase what I think they said (just my interpretation, of course...):
"Only servos which are mechanically connected to the camera module are considered part of the camera module and may be driven by the servo outputs on the camera."
Joe Ross
20-01-2005, 20:41
Ah but by that reasoning what prevents us from connecting the speed controllers to digital IO and using them to generate a PWM signal?
Is there a rule that prevents it? Isn't that basically what PWMs 13-16 are?
Dave Flowerday
20-01-2005, 21:46
Is there a rule that prevents it? Isn't that basically what PWMs 13-16 are?
Remember that PWMs 13-16 though can be shut off by the master processor (even if your code is trying to control them). There is nothing that allows the master processor to shut off the PWMs on the camera, hence hooking anything other than the camera pan and tilt to them would be a major safety concern at competition. When the field crew disables a robot, they need to be absolutely certain that that robot is not going to move and potentially hurt someone in the process.
vBulletin® v3.6.4, Copyright ©2000-2017, Jelsoft Enterprises Ltd.