Scale Model Animation 3 – Teaching The Critter How to Crawl

When last we left off…

When we last visited The Critter, I had tried replacing its power tether with a LiPo (lithium Polymer) battery and discovered to my great surprise that it could be made to go for hours, and certainly long enough to last most operating sessions. This included powering the small Arduino Pro Mini controller with its dual motor drivers. So let’s have a look at that first.

I’ve heard some modelers are scared to touch anything like this controller, or have great fears that they have to learn some deep programming incantations! My answer to that is—Don’t!

You can get this up and running by copying and editing examples in libraries, code (called sketches in the Arduino world) from the web, and my code—which I give to anyone. It’s just not that hard!

Now for a little soapbox speak: So many modelers have claimed we’re a dying breed and younger people are not following along in the hobby. Well… here’s a great example where you can ask a younger party to either help you out or show them your modeling (and animation work!) and they will get hooked!! Are any of you out there aware of the high school robotics groups and their contests all across the world? This stuff is close to that! I HAVE tried this out! (End of soapbox).

To give you some idea I included a couple of pictures here:

The bottom board is a standard Arduino Pro Mini; the top is a home- made driver board with a Texas Instruments SN754410 chip mounted on the underside of the board. The two rectangles in view are 22uf 25Volt capacitors all mounted on a scrap piece of perf board, used for breadboarding circuits. I found this to be of such value that I created a tiny PC board that would mate with the Pro Mini. You can see the rest below:

I use .025 phosphor bronze wire for the pins and low profile machine sockets attached to the underside of the Pro Mini to mate the two. I found a terrific low volume service to make the tiny boards for just a few dollars for 3 boards at OSH Park ( http://oshpark.com/) web service, using the free Eagle printed circuit board design tool—amazing quality for a very low price, with about 12 day turn-around.

Critter in Training…

It was great to get the track drive working and under program control, but the real issue with the animated mobile crane evolved to getting it to “perform!” This meant it needed to do multiple tasks, repeatedly and with precision. So the next step was getting it to go somewhere, accurately. So, you say it needed a human operator to control it?  NO! It obviously (??) needed a guidance system! Now, since NASA and the European Space Agency were already busy, I needed to come up with something myself. So as an (intentional) teaser, what follows is a video of the new Guided Critter making its way through a circuitous, repeatable path. See if you can figure out how it does it! In the next Critter blog installment, I’ll explain how it’s done. It will likely be more complicated to explain it than it was to build it—very straightforward!

Note: there was No human interaction, and No RC control.

Hope you enjoy it and the challenge! The “How” comes next time!

More to come. 

Have fun! 
Best regards,
Geoff Bunza

Magnetic Path underneath.

I'm gonna gander an educated guess that there is some sort of magnetic path/wire under the board that the critter follows.  Much like how Miniature Wunderland animates there vehicles.  I could be wrong!

"Critter" may be following a programmed course

I briefly coached one of the high school robotics teams referred to; the robots compete partially using an "autonomous" mode, during which all functions are pre-programmed by hooking up a laptop to them and loading their memory with specific functions against a timeline.  It's possible the critter has a plug (could that be one on the back?) into which a multiplex wire from the laptop is attached; then, using a specific type of program, the brain can be set to, for example:  MOTOR ONE + MOTOR TWO: forward 5 seconds; MOTOR ONE: stop; MOTOR TWO: forward two seconds:  etc. etc. Using this type of program and some trial and error, a robot (which is actually what the critter now is) can be programmed to follow a certain path, kind of like a cruise missile.  But friendlier!

No, No Programmed Course

If you look at the first crane video:

It does follow a programmed course, as you describe.

The issue with this however, is one of precision and repeatability in the long haul-- friction, and mechanical inaccuracies (not much variation in electrical timing accuracy over the course of a single operatng session on a layout) will cause such a model to eventually go off course. So something needs to keep it on the right path, or if you want, to periodically reset it's coordinates (more difficult I think).

This model actually follows a path, guided.

Best Regards,

Geoff

Guide by wire? Smooth turns...

Dear Geoff,

I'll put a guess in that it's a "follow a wire" type system, the white "tongue"  hanging down between the tracks when the crane is facing the camera is the tell. As to what the tongue is connected to, that detects the "deflection from straight ahead" and commands the appropriate track to move to compensate, now _that's_ where the magic happens...

In a slightly-more visually intrusive form, there are some very cheap toys available which use a pair of IR Tx/Rx sets mounted on the nose of the unit, to follow a black line drawn on a surface. As long as the line is "between the eyes", it goes straight. As soon as one of the eyes "sees the line", it knows that it's wandered off course and corrects. (Interesting sidenote, I did try a figure-8 with one of these, and as long as there was either a "insulfrog" gap in the line at the Xing, or a solid line such that _both_ eyes "saw black" simultaneously, it worked as expected ).

In both instances, a "turn" is a series of "common-direction course-correction judders". I wonder if it would be possible to program the turns so that they were smoother? (Maybe "keep turning until the _opposite_ direction triggers, then correct", with a degree of Kp/Ki smoothing to stop it turning into a "L<> R hunting" movement?).

Love your work Dr Geoff!

Happy Modelling,
Aim to Improve,
Prof Klyzlr

...

The ships in Minatur Wunderland are guided in much the same way their 1:1 counterparts are driven, but local ground stations along the course, that interact with the active beacon on the ship.  Obviously a bit more elaborate than a simple wire track, but still it's possible with the right electronics...some day?

My guesses

Either the aforementioned guide path underneath the platform or something on the surface that is not visible to us (the equivalent of a blue or green screen that the camera doesn't see) but the detector under your crawler does.

Nice work.

I guess the next step would be to animate the boom and cable so as to have the thing do some work instead of joy riding around all day.

Very nice model

My Guess

I also think there is a guide under the board, and on the end of the white actuator there is a magnet. The actuator then follows the guide wire. The actuator  is attached to a device which converts it to a signal (encoder, variable resistor ?) which is feed to the processor and the program instructs which motors to drive to turn, or go straight ahead. But you may have come up with something totally different. 

I really do like your work, new small motors and batteries are opening up so many possibilities. 

rgds

TOMSP

Nice Jeff

The crane follows a magnetic wire. The shoe that you see under the crane as it comes toward you shows that you are using micro switches of some form to activate the proper motor to steer it around the corner.

Here's how they do it in Europe. The use the Faller system.

Nice to see someone is thinking outside the box.

None of the above... Almost!

You people are really observant and smart!  ...But you're not quite there yet !

There are no IR or UV invisible line detectors.

There are no microswitches.

There are no mechanical linkages (like the innovative Faller system).

There are no encoders, no variable resistors either.

BUT, some of you are really close!

Best Regards,

Geoff

All right

Looking at it again I see there are two sheets of either cardboard or plastic. That says to me that there is something between those two sheets. At around 1:20 in the video I see a white show following an "invisible" line.

Sure looks like some kind of linkage to me. Or was this something you forgot to remove before you took the video?

Bernd

Well, Observant but not...

Bernd,

There is nothing between the two sheets of plastic (about .010 each I think by the way). The plastic is on top of a small sheet of foamcore (if that helps-- and it should!). By the way by linkage (mentioned in previous posts) I took to mean mechanical linkage in which case there is none directly relevant.

I hope you all are having some FUN with this! You characters are sooooo close.

Best Regards,

Geoff

Magnetic sensors?

Geoff,

So is it magnetic sensors following an imbedded iron wire or magnetic tape?

Dave

Magnetic sensors?

Soooooooooooooooooooooooooooo close!

So close that you'll argue it's a symmantic difference... but I still can't say you're there.

Best Regards,

Geoff

Dear Dr Geoff, OK, so it it

Dear Dr Geoff,

OK, so it it the difference between

the magnet being on the tongue (mounted on the crane), and the embedded "track" being wire,

VS

the tongue on the crane being "regular wire", and the track being some form of linear magnet (or string of magnets)

????

Happy Modelling,
Aim to Improve,
Prof Klyzlr

For John/Prof

John/Prof,

With a REALLY big grin... No, that's not the difference to which I referred.

Best regards,

Geoff

Some type of sensor

Doc,

Some type of sensor, perhaps a Hall Effect Sensor, and a ferrous or magnetic guide wire/strip under the road surface.

Dave

One last shot

Is it a wire in the foam core that sends out a frequency that is picked up by the magnetic shoe under the crane? Kind of like a phonograph needle works on a record player.

 Bernd

P.S. Does the person that gets the right answer win the crane?

A friend weighs in

Geoff,

I linked this thread to a friend of mine he likes things like this, puzzles that is. Here's his explanation.

How'd he do?

In a phrase: Not even close. Way too complicated for me. 

Bits and pieces of many comments come very close. But you haven't got the right combination just yet.

By the way, there are no RF/signal radiators involved.

Best regards,

Geoff

imbedded magnets

A wild second guess are there magnets in the foam core which repel the magnet on the white arm and so steer the critter within a magnetic boundary.

regards

TOMSP

You are absolutely right!

TOMSP,

You are absolutely right! It was a wild guess! ...and wrong.  

Best Regards,

Geoff

Well, I went back and

Well, I went back and reviewed the video several times, and was sure there is a magnet in the little white thing under the chassis, and it appeared to be tracking a ferrous path hidden below. 

But the last time reviewing the video, I tried listening to the motors for clues, and then I noticed the bird calling in the background.  The BIRD is controlling the crane - I'm sure of it now!  (NOT!)

...I give up.

Whole new meaning

Milt said,

That gives a whole new meaning to the phrase, "Bird Call".

I have to agree with the rest that say there is some kind of tracking system. I'm sure it'll be so simple that everybody going slap their forehead.

Bernd

.