[Dprglist] Compass heading using gyroscopes?
David Anderson
davida at smu.edu
Mon May 18 19:52:11 PDT 2020
And another thing! (picture me with my right index finger raised
significantly, perhaps standing on a soap box)
I also have never had good success with magnetic compasses indoors.
Having said that, odometry is just a way of tracking the robot's
location, and their are other ways. We need a centimeter accurate
indoor GPS, or maybe something with radio beacons?
Karim, I have the same misgivings about robot contests. Mostly because
they tend to focus robot development on very artificial environments,
even though we tell ourselves that somehow these are steps on the way
toward more general purpose robots. In practice the opposite seems to
be true, that the contests year by year produce more and more
specialized robots that can only really run the contest, and can't do
anything else.
On the subject of the four corners calibration, you are correct that the
square doesn't have to be perfect or orthogonal or even really a
square. A rectangle will work. Or a large circle that passes through
the waypoints. Or even a circle that doesn't pass through the
waypoints. As long as the robot in total goes through 360 degrees,
clockwise or counter-clockwise, and stops at what it thinks is the
starting place.
I personally hope that DPRG does not go back to in person RBNOs. I've
enjoyed the virtual meetings and am not likely to make the Tuesday
evening gatherings at the Maker Space. Plus we'd lose Murray, who
probably won't drop by from New Zealand.
cheers,
dpa
On 05/18/2020 04:58 PM, Karim Virani via DPRGlist wrote:
> "Menacing lasers on robots? Hmm, yes", he said stroking his white
> Persian,
>
> Since it looks like the subject is drifting faster than my gyro, I'll
> quickly take one last whack at that dead horse. I never meant to say
> you couldn't get positive results with a magnetometer indoors. I've
> just reset my expectations. I don't have carefully captured data to
> back up my opinion, just anecdotal experiences. I probably tested
> mems magnetometers in fewer than 15 different venues/environments
> before giving up on that line of inquiry. Again, I don't have hard
> data, but my impression is that maybe 90% of the runs I did in those
> environments were anomaly free. And a few of those environments were
> 100% free in the specific paths where I ran robots. I just could never
> predict where I would run into an anomaly or what the
> consequences would be. When thinking toward robots that should be able
> to reliably know where they are over extended duration and without
> babysitting, it seemed to me I'd need different orientation sensing.
> But those experiments did have value for me.
>
> On the subject of contests - I'm internally irritated by them. Both by
> natural and philosophical leanings. I don't get the FTW focus of
> spectator sports much either. But then I come back around to
> remembering that I'm just an outlier primate struggling to coordinate
> with his tribe. Contests do provide the kind of goals and structure
> for progression that my students can organize around. Without
> something like that I'd lose the vast majority of them. It doesn't
> matter that the goals are artificial and it's even essential that
> arenas are protected from real world fuzziness.
>
> Four square does benefit from good alignment to the corner markers -
> though it's not actually necessary. As long as your robot stays
> outside the corner markers as it travels, it doesn't matter if the
> square your robot follows is off a couple of degrees from the markers.
> It just matters how close you end up to your starting point. This is a
> training challenge for my students starting in middle school. It is
> possible to scan with something as simple as a wide beam ultrasonic
> sensor to find the angle to the first marker. But that's because we
> used markers designed to give a reliable signal on a smaller course.
> No such guarantee with DPRG rules. The laser is a good bet.
>
> Rounding up this grab bag of subjects, Texas has just released new
> pandemic guidelines. Apparently youth clubs can open today. I'm
> guessing DPRG doesn't qualify as that .... But my team is discussing
> how soon their parents will start letting them out of the house. We
> might start meeting on Saturdays again soonish. With appropriate
> practices. If DPRG starts in-person RBNOs again soon - please consider
> including the remote component of our recent Tuesday nights.
>
> Forever your most humble servant,
>
> Karim
>
>
> On Sun, May 17, 2020 at 11:24 PM Murray Altheim via DPRGlist
> <dprglist at lists.dprg.org <mailto:dprglist at lists.dprg.org>> wrote:
>
> Hi Doug,
>
> Thanks for the info on the smaller LiDAR and the YouTube video,
> which I just
> watched. If I'm trying to measure distance, the Benewake products
> look really
> good, especially that smaller one.
>
> The idea of measuring a gap was a possible approach to aiming at
> the first
> target of the four square challenge. It'd be kinda cool to have
> the robot be
> able to locate that gap at the beginning of the course rather than
> rely on
> the robot's owner physically aiming it at the target, but either
> works: I
> like the complexity of the LiDAR and the simplicity of the laser
> pointer.
> Hacking a laser distance meter might work but it's pretty big
> physically for
> my somewhat-small robot. By contrast, the tactical red dot laser
> I've just
> ordered from somewhere in China is 42mm x 20mm x 24mm and weighs
> 28g. That's
> small enough to bury under my robot's bumper. For the same price
> (US$13.55)
> I had the choice of a class 2 or class "IIIa" ("less than 5mW")
> laser and
> bought the more powerful one, which might be in the cat-tattooing
> power range.
>
> Currently I've got that little Pimoroni VLX53L1X Breakout Garden
> sensor with
> a range of up to 4m (though it seems more like 2.5m in practice,
> really) and
> is about the size of a postage stamp (remember them?!). I've
> mounted it on a
> micro servo and have a draft post on my blog that I've not
> finished yet;
> suffice it to say that it's a very capable sensor and can do an
> accurate 90°
> sweep in about 3 seconds. The power requirements are almost
> inconsequential.
>
> David was asking me about it on our last video conference. If you
> mount an
> ultrasonic sensor on a servo and try sweeping it that quickly you
> get nothing,
> sound can't travel that fast, but the little laser works really
> well (though
> yeah, the speed of light ain't what it used to be).
>
> Another idea I had was to use one of the Raspberry Pi cameras to
> look forward
> towards the target and locate some specifically-colored pixels.
> The resolution
> on the cameras is probably high enough that this might work if the
> lighting
> in the room is good enough. Or, aim the camera at some kind of
> point light
> source. The rules permit something like that.
>
> But if I'm not trying to measure distance, simply point at
> something as a
> target, a small tactical red dot laser (with allen screws to
> adjust trajectory)
> is just the thing.
>
> I appreciate the point of conversations like this is to "advance
> the technology"
> that could be used on our robots. So using a LiDAR rather than a
> cheap laser
> pointer would be a greater advance and a more interesting
> experiment. So I'll
> continue to think about this (even though I've got a menacing
> black laser on
> its way in the mail).
>
> Cheers,
>
> Murray
>
> On 18/05/20 2:57 pm, Doug Paradis wrote:
> > Murray,
> > I don't know what size gap you are trying to detect or the
> speed that> the gap may appear. Hhowever if the sampling rate is
> slow, about once > per second, another possibility is a laser
> distance meter with a serial> port. You can get a
> laser distance meter board with a serial port from> Aliexpress (I
> have seen them listed, but can't seem to find the right> search
> terms, so no link), or you might look at this project
> >
> (https://hackaday.com/2014/03/29/hacking-a-laser-tape-measure-in-3-easy-steps/).
> > I own one of these meters, have brought the serial port out, and
> played> with it. Its cool, but a robot generally needs a faster
> update rate.
> >
> > Regards,
> > Doug P.
> ...........................................................................
> Murray Altheim <murray18 at altheim dot com> = = ===
> http://www.altheim.com/murray/
> === ===
> = = ===
> In the evening
> The rice leaves in the garden
> Rustle in the autumn wind
> That blows through my reed hut.
> -- Minamoto no Tsunenobu
>
> _______________________________________________
> DPRGlist mailing list
> DPRGlist at lists.dprg.org <mailto:DPRGlist at lists.dprg.org>
> http://lists.dprg.org/listinfo.cgi/dprglist-dprg.org
>
>
>
> _______________________________________________
> DPRGlist mailing list
> DPRGlist at lists.dprg.org
> http://lists.dprg.org/listinfo.cgi/dprglist-dprg.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.dprg.org/pipermail/dprglist-dprg.org/attachments/20200518/b13912c2/attachment-0001.html>
More information about the DPRGlist
mailing list