Shop OBEX P1 Docs P2 Docs Learn Events
Basic info requst — Parallax Forums

Basic info requst

Maxan3dMaxan3d Posts: 4
edited 2005-12-10 16:01 in Robotics
Hi,

I'm a 3d graphics artist and freelance match mover for video/film effects work. However, I have NO idea where to start on this project. I want to make a motion control rig for light-weight miniDV cameras. Now, I know I could buy a Scribbler robot and use the parts to create simple dolly rig like for repeating the same dolly movements over and over. But, I want to make a complete rig using programable component parts that aren't from "toy" robots for durability and steadiness.

Here's my question: I "think" I can get some servo motors, an MMC and Stamp controller to make what I want. Can I use my PC to program this Stamp component to repeat a specific motion with any precision?

You can probably tell from this post that I have no clue what I'm talking about! [noparse];)[/noparse]

ANY info would be greatly appreciated. I have no idea where to start other than getting the Scribbler and starting from there.

Thanks,

Dave
www.max-realms.com
www.max-animation.com

Comments

  • allanlane5allanlane5 Posts: 3,815
    edited 2005-11-14 14:26
    The timing on a BS2 may not be repeatable enough for what you are trying to do.

    As I understand it, a motion control rig MUST move the camera in EXACTLY the same path several times -- once for the background, once for the spacecraft, etc.

    The BS2 (and the aircraft servo's that it supports) is kind of 'sloppy' -- you get 2000 Basic Instructions per second (about 300 uSec per instruction, actually) and I don't know how 'repeatable' a servo position is. And the 'clock' in the BS2 is 'resonator' based, so it's +- 1% accurate.

    You COULD use the BS2 as a 'brain' for more precise components, though. Have it issue SEROUT commands to another system to move stuff. The repeatability is there in the program, the question is the 'slop' in the hardware components.
  • Maxan3dMaxan3d Posts: 4
    edited 2005-11-14 16:24
    Thanks man,

    That's pretty much what I had suspected. Stumbling into the field of robotics in my research has been like falling asleep and waking up in Paris. I don't speak French.

    Thanks for the help.

    Dave
    www.max-realms.com
    www.max-animation.com
  • Jon WilliamsJon Williams Posts: 6,491
    edited 2005-11-14 17:08
    In addition to my job at Parallax I'm a professional actor (see www.jonmcphalen.com) and shot/edit a bit of video.· I, too, am interested in this kind of project but I think that it will take a bit more than servos to do it.· I have thought about building a robot pan/tilt head form my camera using stepper motors (precise and repeatable) and proper gearing for torque.· Of course, it's best to to have the camera balanced on the head just as a camera operator does with a big Panavision camera on a standard pan/tilt rig used in film.

    The BS2 would work fine for control; use steppers and proper gearing for mechanical precision.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Jon Williams
    Applications Engineer, Parallax
  • Maxan3dMaxan3d Posts: 4
    edited 2005-11-14 17:28
    Thank you Jon. You guys have given me enough to continue researching in the right direction. [noparse];)[/noparse]

    Dave
    www.max-relams.com
    www.max-animation.com
  • Ken GraceyKen Gracey Posts: 7,392
    edited 2005-11-14 18:36
    Dave,

    I've seen too many users of microcontrollers discouraged by experienced people who tell them they need interrupts, background processing, lots of RAM, etc. Look through the trees to see the forest and start by learning. The cheapest way to experiment and gain a basic understanding of the Stamp is to go to RadioShack and pick up a What's a Microcontroller Kit. For $79 you'll be on your way.

    The reported "sloppyness" of BS2 timing may not be an issue at all in this case. You will need a stepper motor driver, and I suggest picking one with intelligence. Dealing with mechanical issues is enough on it's own, and the BASIC Stamp should make your electronic control system a bit easier.

    Most microcontrollers can do what you want to do and many of them require you to connect different helper devices to them to manage the motors. Even if you can do it all "with one chip" you might not be interested in the cost/effort ratio. Sometimes $79 for a Little Step-U or even $129 for a Gecko G201 will go a long way to make life very easy for you. Rather than driving the steppers with pulses from a BASIC Stamp, you might send a serial command identifying the motor and steps to take, or a couple of pulses signifying the step and direction. The BASIC Stamp is fairly expensive, but it's easy to use. And for somebody getting started you came to the right place.

    My suggestion: quick trip to RadioShack. Just dive in.

    Ken Gracey
    Parallax, Inc.
  • Maxan3dMaxan3d Posts: 4
    edited 2005-11-14 18:42
    Hi Ken,

    Thank you for that. My wife's getting ready for our trip to Radio Shack as I type this message. You guys have been a great help and I really appreciate it. I'll be sure to come back to this forum and let you guys know if I can figure anything out!

    Dave
    www.max-realms.com
    www.max-animation.com
  • Tom WalkerTom Walker Posts: 509
    edited 2005-11-14 21:11
    To add to what Ken posted...unless you are doing VERY high speed shooting, duplicating camera moves is probably going to be more of an issue for your mechanics, precision and backlash being the real problems. As Jon pointed out, once you have your camera well balanced, being precise in space is mostly a matter for the repeatablility of your mechanical design.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Truly Understand the Fundamentals and the Path will be so much easier...
  • J. A. StreichJ. A. Streich Posts: 158
    edited 2005-11-17 08:11
    An intresting idea...

    I'd like to add to it. I can't see programming the camera "go 10 feet, zoom, go 10 more while turning 30 degrees..."

    If you are able to get the repeatablility you may want to consider including an RF receiver, record the data that comes from it, then replay it... If there is no input from the RF, then do it exsactly like the last, but if there is input make the small adjustment (or complete adjustment) to the pre-recorded commands. That way it's a lot more smooth than sheer numbers at the computer (away from "the set").
  • kelvin jameskelvin james Posts: 531
    edited 2005-11-19 07:34
    A robo-dolly, sounds interesting. As you probably know, the biggest issue in camera movement is keeping it smooth and stable, which at times can be a problem. Panning is usually done very slowly to eliminate any jitter that can show in the frames, more so with DV. So, when putting something like this together, having a smooth, fluid motion will be the major concern. I had built a related project to use on a DV camera, using a stepper driven system, and it works good. But, at very slow speeds, it does in fact "step", which in your application, will probably be picked up in the footage. My controller will only output half-steps, but a micro-stepper would probably cure it. Other than that, the positioning and speed, is right on for repeatibility. The higher resolution motor you can get, the better, a .9 degree would be ideal. Mine can move through multiple positions, at different speeds and times from all the recorded points. The thing does work, but i am not entirely happy with it, so i am looking at going to a servo/ encoder positioning system, that will give a more fluid movement. Anyway, if you need some guidance, i'll try to help along the way. Are you going to do this on wheels or a track?

    kelvin
  • LoopyBytelooseLoopyByteloose Posts: 12,537
    edited 2005-12-10 16:01
    If you really do need more speed than the BasicStamp, you can move up to the SX-28 and program that in SX/B which is another form of Basic particular to the faster speed of the SX-28.

    Get started with the BasicStamp.
    You can use it to gain solid insight into where you want to go.

    If and when you move up to the more sophisticated SX-28 or SX-48,
    you can also return to it to get new projects started.

    In other words, most of us find that the BasicStamp is very handy for design and decision testing. Even when we need something more to get the final project done.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    "When all think alike, no one is thinking very much.' - Walter Lippmann (1889-1974)

    ······································································ Warm regards,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan
Sign In or Register to comment.