SpudFiles

You still don't get the fundamental problem. You'll just have to come to that obstacle yourself I guess (If you ever do any real development).

Teach me sensei!! Teach meeeee!!

I've seen it. It is just as bad as the others. I've seen them all and this one is clearly the most advanced by a wide margin.

Forgive me sensei I forgot that you're not advertising your own product

Somehow I am more impressed by the one built by a teenage boy than yours... he doesn't charge for the software, his is opensource and he has less resources to spend on it
(not to mention that he IS a teenage boy not some 20-35 yr old dude who pretends to be someone he isn't)


lol stop... this is getting more and more hilarious

JeffSinger wrote: You can't really trust the software to keep accurate track/time of stepper movement. CPU load is usually maxed out in image/video processing. When that happens some timers (threads) stop all together. I know what advantages steppers could have. The problem is that there isn't a stepper out there that you can just give a position command to like an RC servo. You'd have to build some sort of RC servo-stepper mutation where you somehow took the PPM board and POT from a servo and mounted it to a stepper. You'd have to create a controller that was a mix of stepper and servo so that you fed it position commands instead of steps and instead of sending out DC current to a regular DC motor, it controlled the movement of a stepper. As far as I know, nothing like that exists.

Like I said before, stupid in is stupid out regardless of the hardware.

Any real image processing app is going to use dedicated hardware for the job. It would at least be done on a GPU. That leaves basically the entire CPU available for the motor control.

Besides, motor control on a modern CPU is only going to need something in the vicinity of 1% of the CPUs time to deal with the motor. If your app is that CPU bound that you can't spare the 1% for decent motor control then stepper vs. servo is the least of your problems.

Heck, in many robotic apps the motors are only updated once every millisecond or so, which is still at least an order of magnitude faster than they really need to be. Even running a pid the math to deal with the motor can be done in less than a microsecond.

POLAND_SPUD wrote:

I've seen it. It is just as bad as the others. I've seen them all and this one is clearly the most advanced by a wide margin.

Forgive me sensei I forgot that you're not advertising your own product

You're making several assumptions to come to that conclusion. You know what they say about assuming...

jimmy101 wrote: Like I said before, stupid in is stupid out regardless of the hardware.

There is a fundamental problem, and then several other smaller problems with steppers. I know what you're saying but I'll believe it when I see it. As common as it is for people to think steppers are the way to go I'm betting there have already been several attempts at this with steppers. You don't hear about them because they failed for the reasons I explained before.

I doubt any serious attempt to make a sentry gun with steppers would fail because of the use of steppers.

Near as I can tell, servos are almost always used in low power applications. For applications that require serious power from the motor(s), steppers seem to dominate.

jimmy101 wrote:I doubt any serious attempt to make a sentry gun with steppers would fail because of the use of steppers.

Near as I can tell, servos are almost always used in low power applications. For applications that require serious power from the motor(s), steppers seem to dominate.

The reasons I mentioned several posts ago are why they aren't used.

well I might be new to electronics and stuff but I think that instead of tracking the exact position of a step motor (or a 'true' servo with encoders) it would be a lot easier to mount an IR laser on the gun and use the camera to track it.
That seems a lot simpler... sure there is a risk that someone might use this to jam it but I guess there is a way to fix that.

I wonder if it would be possible to build a simple range finder based on this [youtube][/youtube]Though that would require an aditional camera (mounted on the gun)

@Jeff
Since you already have some experience in removing IR filters from cameras you might as well give it a try

POLAND_SPUD wrote:..it would be a lot easier to mount an IR laser on the gun and use the camera to track it.

nope. I've already played with that method. One of the reasons it's no good (out of several) is because of the slight lag between real-time and the time that the PC has finished processing the frame. It is fine as a slow range finder though as long as the PC can see the dot. Won't work outside without a much more powerful/expensive laser.

One of the reasons it's no good (out of several) is because of the slight lag between real-time and the time that the PC has finished processing the frame

that seems kind of counter intuitive - the software can detect motion and estimate lead but it cannot track a small dot?

Hmmm I found processing code that does this but the only webcam that I have is on the laptop... so that would have to wait...

What about the IR sensor on a Wii remote? It seems to be calculating the position in near real time...

Anyway you've said that steppers occasionally miss steps... now you've got a simple method to zero the turret...

Steppers do drop steps if run beyond their design speed / load limitations. When run within their design limits, they are very reliable.

POLAND_SPUD wrote:

One of the reasons it's no good (out of several) is because of the slight lag between real-time and the time that the PC has finished processing the frame

that seems kind of counter intuitive - the software can detect motion and estimate lead but it cannot track a small dot?

Hmmm I found processing code that does this but the only webcam that I have is on the laptop... so that would have to wait...

What about the IR sensor on a Wii remote? It seems to be calculating the position in near real time...

Anyway you've said that steppers occasionally miss steps... now you've got a simple method to zero the turret...

I could sit here and point out why you are wrong on all points but that is a lot of work. It is better to just let you discover those reasons yourself when you try to make your own.

velocity3x wrote:Steppers do drop steps if run beyond their design speed / load limitations. When run within their design limits, they are very reliable.

Missing steps is not the main issue at all. See my previous posts.

JeffSinger wrote:

velocity3x wrote:Steppers do drop steps if run beyond their design speed / load limitations. When run within their design limits, they are very reliable.

Missing steps is not the main issue at all. See my previous posts.

I didn't state that missing steps is the main issue. I'm quite familiar with the stepper / servo controversy. I have many of both. Several of my steppers are 2,200 in/oz and run on regular basis. I was pointing out that steppers are very obedient motors in certain applications. Steppers being an open loop system are the wrong choice for a sentry gun but, they have their place....... OK?

well whatever... I see that your company is doing relatively well ATM.... but one day you'll have to improve it, whether you like it or not...

...and the excuse 'RC servos are the bestestest' won't work... that hasn't convinced me and the others

your setup is just one tiny step ahead of the one built by that teenage dude (mostly due to the fact that you're like 2 times older than him and you have spent more on it than his pocket money allowed him)... and it's lightyears behind more professional stuff... so either you're going to acknowledge that fact or you're scre###

I am new to this stuff so I might not be the best person to tell you that...

but from my perspective.... in the past your turret looked pretty impressive, 'I was like hmmm I wonder how the board works?' now I am more like 'dude your charging $X for a thing that controls 3 servos, WTF ?!?'

I don't really need an autonomous turret... remotely controlled one is all I need... mostly due to the fact that I might actually put something on it that is potentially lethal... but if I ever wanted to build one I just have to copy paste the code and it will work... if I wanted I might even change it to suit my need better

For me (and probably a lot of other ppl) an autonoumous turret is just a cool gadget... you build it, you test it, have fun with it, and then you forget about it....
arduino is perfect for that... I build a turret with it - when I get bored with it I'll build something else with it...

yours costs almost 300$.... so you haven't convinced me that your project is worth the price... and you probably won't convince others

so my advice is -> realize it

You must be trollin. Either that or you have a lot to learn.