White Box Robotics Pre-order

To a growing number of people, reading an analog clock is becomming more difficult. In any given population, only a certain percent of them will be able to program a VCR.

The issue isn't reaching out to a broader audience, per se', but being able to indentify a large enough audience to make it worth while. I like Einstein's quote: "It should be as simple as possible, and no simpler."

Hopefully the person building the robot knows enough to think for themselves, hense the part in the above quote: "... and no simpler."

Again, on the site: "a little skill, and some good old fashioned scrounging." should speak volumes.

You can't make a product aimed at everyone unless it does nothing. A pet rock for instance, has a very broad appeal. As you add specifying features, you narrow your market. It just so happens that the hobby market is pretty specific to people who like to build things.

Out of 1000 people interested in building a robot themselves, how many do you think would have a problem building something out of angle aluminum and toy car wheels? If you say 10, that's 1% and not a problem. If you say 100, that's 10% and still not a real problem. If you say 350, that's 35% and still not a problem. If you say 800, then that's 80% not a problem if the overall population of the set is a few million.

You can't market to everyone, you have to identify your audience.

That is a choice people will make.

I do, and I'll tell you why. You and I could probabely build the equivilent device for a similar amount of money. That is a bad starting point for a consumer item. When you deliver a product to consumers, it should be something better than they can build for themselves and far cheaper.

Do you have *any* idea of what it would cost your or I to hand wirewrap a PC motherboard? Buying a PC motherboard is a no brainer because the economies of scale have made it much cheaper.

The whitebox robot offers no real advantage for the experimenter. It may save some time, but it doesn't save any real money, and to the hobbiest, it isn't "thiers." Don't underestimate that hobbiests want the pride of accomplishment.

True, but an easly remedied sort of thing for a hobbiest. One of the things I want to stress on the web site is that this is a minimalist sort of thing and that the builders should feel free to improve at will.

"Most" of whom?

Have you ever watched cooking shows or automotive shows? There is an expectation of a certain level of pre-existing criteria. The $500 dollar robot is just such a project.

Again, "This old house" assumes you own the tools, doesn't it?

and my $500 robot is an ideal, a target. When I've found and identified all the peices that bring it in under $500 then I'm done. Does that mean you wouldn't want to make it better? Does that mean it is not a good source for code and techniques?

It isn't so much a "Build *this* robot" as it is "build *your* robot" project.

Reply to
mlw
Loading thread data ...

cheap ultrasonics: Polaroid cameras. Thrift store and swap meet.

Bumper system: microswitches from keyboard, junk toys/radio etc. front bumper: 6 switches, with 3 bumbers. Coat-hanger wire between two switches:

/-----\ /-----\ /-----\

  • * * * * *

you can detect obstacle at 11,12, and 1 o-clock.

Best done on all sides, for total of 24 switches.

Rich

Reply to
aiiadict

Okay, but what happens when you want to use the robot at some other time, say 7 o'clock?....

-- Gordon

Reply to
Gordon McComb

Ok,

compass directions.

ne, n, nw

If you want to use at 7 oclock better have it running autonomous.

any time after 6 is time for beer, and you don't want a RUI.

roboting under influence.

Rich

Reply to
aiiadict

Or getting the beer. Still working on that one.

-- Gordon

Reply to
Gordon McComb

This is my point, you seem to look at the people in this hobby as all wanting to *build* a robot. Many do, many don't. I've been in a few robotics clubs and have lots of engineering friends (mostly software) and have met several people that are not interested in building the actual robots. They just want to program a robot to do cool things. Some don't want to build the robot because they don't have intrest. Some don't want to because they don't have the time.

Are you saying that if someone doesn't want to build a robot, then go away? Compaines shouldn't make robots like Whitebox's?

You think so? Are you talking the whole Whitebox robot, just like you see it on the site? I recently started a business building a consumer electronics device and manufacturing a complex device such as this is not cheap. I would really like to see you build the same thing for $1200, not even counting the # of hours designing and building it or tools needed (which you apply no value to).

Thats not a valid comparison in this instance.

Again, this is showing your closed mindedness to the people that make up this hobby.

"Most" was a poor choice of words. Meant "many". People that I describe in the paragrapgh following this sentance.

Actually, in many of the shows I watch (I'm very much into cars as well) they are constanly explaining the tools, where you can get them and how much they cost. Many times they proceed the show telling you about special tools you need for the project and their costs to buy or rent. Most the shows I watch also start off by telling you how much time will be invloved.

Again, they explain the tools and their costs all the time. They constantly explain the cost of hiring someone for specific part of the project, various tools and rental costs, etc. They don't ever base the costs of a project on the materials alone.

By "ideal" you mean starter? My argument is nothing about whether your robot would be a good source for coding or not. Its that you dismiss the Whitebox robot because you think it doesn't have a place just because *you* where able to build something similar for less. You seem to assume everyone wants to go the same path as you. That they even have the time to do so. You put little to value in the amount of time involved, the *want* to do this approach, and having the tools and space to do so.

-C

Reply to
Hoss

I actually have a ton of old polaroid ultrasonic ranging systems that are 20 years old. They have the old gold-foil transducers.

I want something that is obtainable by others.

I was thinking of using a keyboard, but I have three switches available through a mouse that I may want to use. That way, they would go directly to the motor control software.

Reply to
mlw

formatting link
Mitch

Reply to
Mitch Berkson

Hey, I never said they "shouldn't," I have said that I doubt it will fill a viable market segment.

Assume these characteristics: The build it yourself crowd won't pay much money, they want something cheap on which they can build, or build it from scratch. Does that seem like a fair statement?

The "buy" it crowd, unless directly employed in the field, will only buy it if it is within their "hobby budget" or desposable income level.

Either way, unless it fills a specific need that adds value beyond the hobby, you are bumping up against the customer's descretionary spending limit. $1000 is a lot of descretionary spending cash.

There is a difference between building something for a specific task and something for education or hobby interests.

Hint: when debating a topic, debate the topic, not what you think the other person is thinking. The facts stand for themselves, bringing in my, or any others supposed personality traits, is an ad hominem attack and shows that you have to critisize a person because you can't discuss the topic.

In short, your sentence was intended to insult and discredit me thus attempting to indirectly discredit my opinion.

Whether or not you think I am "close minded" is irrelevant. I still assert that hobbyists want the "pride of accomplishment," and unless you can post something to refute this statement, it stands regardless of your assertions of my supposed closed mindedness.

Many is a relative term, and sometimes a few of something is enough.

Obviously the expain "special" tools, by the assume everyone has screwdrivers and wrenches, true?

Again, out of the ordinary tools, obviously.

"This Old House" no, but many of the improvement shows, including custom car shows, talk about "weekend rebuilds," or "things you can do for under $$$"

I mean, low entry cost. It takes very little to get reward. The "instant gratification" society.

ok.

NO, I think I've articulated why I don't think it will work, it isn't about my, it is about how I think I see the market.

Not at all.

That sort of is the hobbyist market, is it not?

Reply to
mlw

The Polaroid ranging systems are easily obtainable by others, either in kit form, pre-built, or scavenged from old cameras still widely available through eBay (though it's been a while since I looked) and as the previous poster mentioned, flea markets and swap meets. In the case of the latter, there is plenty of information on the web on hacking these old camera boards and transducers for use in robotics. Zagros, Acroname and many others sell the former as either kits or assembled units.

Devantach also makes a number of models of inexpensive rangers. The I2C-enabled versions are more expensive than the "time the pulse yourself" versions, but could probably be interfaced to your linux box via the SMBus, and will compute the TOF on your behalf. Most of the Devantech models have a wider effective FOV than the Polaroid units, a lower minimum range and a lower maximum range. This makes them maybe a little less well-suited to distance mapping, but great for obstacle detection and close-in mapping. They also tend to be easier on the power supply than the Polaroid rangers. I generally prefer these.

-- (Replies: cleanse my address of the Mark of the Beast!)

Teleoperate a roving mobile robot from the web:

formatting link
Coauthor with Dennis Clark of "Building Robot Drive Trains". Buy several copies today!

Reply to
the Artist Formerly Known as K

Actually I don't think anybody claimed that you *couldn't*, just that it is easier to do on a microcontroller.

Frankly you could do your PID calculations 1/second for the most part and get it done nicely.

My problem is that your motor controller seems like a Rube Goldberg device.

You have a DAC on your PC to create a voltage, which then goes into a controller which converts that into PWM to power your motor.

To me it seems *much* easier to buy a commercial motor controller. Or even to make one yourself. You can program in C.

I'm not saying that you *can't* do it: you've proved that your system works. I'm just saying that there are cheaper and easier systems out there for motor control.

I've seen people in rattlesnake bagging competitions and people who do bee-bearding. Not things *I* would do, but they proved it could be done.

Good luck to you. To me this seems to have been done many times before.

-- D. Jay Newman

formatting link

Reply to
D. Jay Newman

Hint: when posting to Usenet, try not to come off as monumentally arrogant when expressing an opinion. That way, you won't feel the need to defend yourself against "ad hominem" attacks every other post. This is not to say that you ARE monumentally arrogant -- I don't know you personally -- but you come off that way.

Hint: there's no need to spend three paragraphs defending yourself against "ad hominem attacks". It's undignified. Plus, until you've gotten into a discussion of any length with R. Steve Waltz, you really don't know the meaning of "ad hominem attack".

Hint: grow a thicker skin -- it's Usenet.

Bonus Hint: don't start a paragraph with "Hint:". It's condescending and makes you come off as monumentally arrogant. Only I (and I alone) can get away with this.

Reply to
the Artist Formerly Known as K

I only "come off that way" to people who disagree with me but do not have the background to refute. :-) There is a lot of psychobable about why people may feel the way they do, but it always has to do with their own issues. I agree with Eleanor Roosevelt, "No one can make you feel inferior without your consent."

You are discounting the fact that I may enjoy writing.

I do not know Steve Waltz, I guess I should be thankful. I always did well in debate at school and really find personal crap in a topical discussion irritating.

My skin is thick enough, it is just a lecture about "ad hominem" attacks or respond in kind, and I'd rather not stoop to that level.

You fail to recognize that this was the intended tone. The point had to be made. When people forget they are discussing a topic, and bring their person feelings into it, it makes for a crappy discussion. Why not just bring in "you're so dumb..." jokes.

Hint: Not only you :-)

>
Reply to
mlw

And, without opening that can of works, is a difference in perspective.

Well, that only works if you are controlling something of sufficiently slow response times.

Oh, man. That's just mean. :-)

This is a fairly standard methodology in a lot of industrial controls.

Again, we don't want that argument again, do we? We can both state a viable case where we can prove our own opinions.

Stop it, I'm laughing too hard.

Actually, it may have been done before, but I guess the thust of my project is to highlight the difficulties and the possible solutions. For instance:

int PIDControl::CalcPID(double target, double actual, double scale) { // Calculate the physical error scaled to compensate for // fluctuations in time base. double error = (target-actual)*scale; // Calculate the differential between last error from scaled values m_differential = error - m_error; // Update integral with the error (should average out) m_integral += error; // Save the last last error m_error = error; // Sum the raw PID components using the scaled values double rawpid= (error*m_gain_error)+(m_integral*m_gain_int)+(m_differential*m_gain_dif); // Adjust for bias int pid = (int) round(m_bias + rawpid); #ifdef TRACK_PID if(m_logfile) { fprintf(m_logfile, "%f,%f,%f,%f,%f,%f,%d\n", target,actual,m_error,m_integral, m_differential, scale, pid); } #endif // Return clamped PID value return (pid > m_pidMax)?m_pidMax:(pid < m_pidMin)?m_pidMin:pid; }

In the above example, the "scale" parameter is absent in virtually every PID code I've ever seen, but it makes a difference in motor stability -- even on highly consistent system.

Reply to
mlw

have you ever tried to design a PC with no external microcontrollers that does:

PID voice recognition streaming video to net auto-navigation voice synthesis sonar sensors bump sensors arm control stereo-vision

3d modeling of environment object recognition landmark recognition ?

I would love to see how you get all of this stuff done and not bog your PC like your above mentioned windows-running-many-time-critical-programs scenario.

Rich

Reply to
aiiadict

Did you not read the post to which you are responding?

You accuse me of misrepresenting what people have posted, when you are probably one of the worst culprits. I have never said micro-controllers do not have a place, never once. I have said that I have no interest in developing micro-controller code and have no desire to use micro-controllers in a capacity where their function is not well defined. My phrasing may have been different in different posts, but the intent has been consistent. You have taken this to mean that I don't want to use any micro-controller based technology, which isn't true. A self-contained stand-alone device whose behavior is predictable, well defined, and cost effective would be acceptable (linksys router, hard disk, etc.) Again, I've said as much in other posts.

Reply to
mlw

If you would characterize linksys router firmware as "predictable", I shudder to think of what you'd consider non-deterministic...

Reply to
the Artist Formerly Known as K

At best this is an evasion. You well know that everyone else has been talking about a microcontroller you program yourself, as opposed to something that comes in a commercial product. Get real.

You keep saying you'll a use microcontroller if its function is well-defined, but you have not said why a custom-programmed microcontroller -- of the type EVERYONE ELSE has been talking about -- cannot have a well-defined function. I can only take that to mean a function *defined by you*, which makes it a guessing game for the rest of us. What fun!

No one is trying to put words in your mouth, but when you don't explain yourself well, this is what happens.

BTW, get a spell checker. There is no such word as "hobbiest."

-- Gordon

Reply to
Gordon McComb

You assume a lot, more than your experience in these matters should warrant. You're not yet the expert, so stop pretending you are. Both of your assertions are meaningless because they are your perspective only. If you want to state opinions, call them opinions. If you want to state fact, get some. Do some market research.

After you do your research, you can explain why there are $800 hexapod walker kits, and $100 hexapod walker kits -- and a bunch in between these price points. They've all been on the market for quite some time; they are selling well enough for continued manufacture.

No, he's making an observation, and not a unique one at that. Several others have made the same. He's not calling you stupid (that would be an ad hominem attack) but saying you are closed minded to the people that make up this hobby. Since he's one of those people, it's a valid statement.

Why should anyone try to refute this? You've just invented "pride of accomplishment" as your argument. Out of just one sentence you previously said, no less!

Hey, you're starting to spell hobbyist correctly now!

-- Gordon

Reply to
Gordon McComb

Nah. This isn't an explanation, but another evasion. The behavior of tested code in some microcontroller is also understood and, especially when source code is provided, well understood.

Try again.

You have to explain it a number of times because you didn't explain it well the first time.

Oh. I was just trying to help you refine your writing, seeing how you want to contribute to magazines, or write a book. I thought you'd want to know. My mistake.

-- Gordon

Reply to
Gordon McComb

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.