[ARDF] Looking ahead to ARDF apps (in 2012)

[Charles Scharlau]

A digital compass like those used in mobile devices are a different beast
from the common mechanical bar-magnet compasses we're all used to . Both
are subject to errors from various causes. But by their different natures
they require different treatment in order to get good accuracy.

Smart phones (Android, iOS, BlackBerry, etc.) all use 3-axis digital
compass modules, and 3-axis digital accelerometers, in unison to generate
tilt-compensated magnetic compass data. It is quite remarkable how well
their compasses work in any physical orientation, even holding the phone
vertically or upside down. Try that with your average bar magnet compass.

But as you note, digital compasses are subject to losing their calibration.
And when they lose calibration, they are subject to providing ridiculously
large errors, even showing 180-degrees opposite what they should!

But I've found that it is possible to get good reliable magnetic bearings
from these mobile devices. Possibly Apple has introduced some improvements
in their compass algorithms with iOS 4.x and 5.x, since it seems to me that
reliability has improved with time on our iPhone 3Gs.

Also, I found that when I stopped carrying my mechanical compass with me
that the cell phone compass instantly began performing much better. The
strong magnet in the mechanical compass was wreaking havoc on the cell
phone's compass calibration!

I've found that the following rules, if followed carefully, allow my Apple
products give good bearings:

1. Keep the phone away from strong external magnetic fields. That includes:
mechanical compasses! car hoods and roofs; wires carrying currents (e.g.,
under the dash board); ear phones, other cell phones, or anything with a
speaker; fingernail clippers, scissors, knives, or anything made of a
ferrous material. Try to keep the phone at least 1 meter away from any of
those things after it has been calibrated.

2. Carefully calibrate your phone before trying to use it for
heading/bearing measurements. Smart phones will calibrate their compasses
automatically, and continuously. They can even determine when their own
calibration is seriously out of whack, and they will prompt you to do the
"magic wave" when it is required. But you should do that wave at least once
before using it, even if it isn't prompting you!

3. Calibrate by slowly rotating the phone 1080-degrees (3 full rotations)
in all three rotational directions. The phone must be turned on at the
time, and held away from any strong external magnetic fields (see 1 above).
Don't even think about calibrating it in the car.

========

The great circle calculations used in the app you sent me must have been an
early version. They appeared to be an attempt at implementing the following
formulae in C:

http://williams.best.vwh.net/avform.htm

The app contained several errors that prevented it from giving reliable
results. I restarted from Mr. Williams' formulae which are used in many
calculations within my app. If you're interested I can provide you with the
Objective-C class I created for great circle math - it works well.

=========

The web site you reference from 2009 describes a raster map web
application, requiring internet connectivity and a remote server. It states
"By starting to develop the solutions now, we’ll be ready when the
technology comes along in a few years time. The solution described in this
article is to be considered as “proof of concept”"

A few years from 2009 is today! And I think the app I've created for iPhone
is approximately what that World of O mobile phone article was looking
forward to.

This app is (almost) all you need for orienteering (and ARDF) right on your
cell phone or tablet: the map, the compass, the course, and the ability to
record your track and bearings. You can hold it all in one hand. No
internet. No 2-year contract. Not even a SIM card is required.

73,
Charles
NZ0I




> Message: 1
> Date: Fri, 30 Dec 2011 15:59:52 -0500
> From: Vadim Afonkin <vadim.afonkin at gmail.com>
> Subject: Re: [ARDF] ARDF Digest, Vol 100, Issue 2
> To: ardf at kkn.net
> Message-ID:
>        <CANxYhxj=fL9RR7ex1k=o0NEZKTxk1QUqdtbv7PLCjcbU4N6fPQ at mail.gmail.com
> >
> Content-Type: text/plain; charset=ISO-8859-1
>
> I made a similar iPhone application about a year ago - it was taking a
> bearings and calculating fox locations, putting it on the map and even
> could guide you to the location, telling how far you are, etc., but I have
> abandoned project because iPhone has a such poor quality compass - it can
> give you errors up to 90 degrees.
> I came to conclusion if I want to get a good application with precise fox
> location calculations and guidance I need to use external compass.
> I made one and testing it. The errors I am getting with small external
> compass is only +/-2 degrees. You really need to use good tilt compensated
> and compact compass. I am using HMC6343, it is a best IMHO. Not chip but
> nice - all in one - $95 from DK.
>
> Anyway - I am a big supporter of this idea. I believe I already sent you C
> code to calculate fox locations for iPhone.
>
> Regarding Orienteering - you can get all you need for orienteering on
> iPhone and Android from here:
>
> http://news.worldofo.com/2009/11/19/orienteering-map-gps-part-iii-mobile-phones/
>
> I tried it on iPhone and it worked for me.
>
> 73!
> Vadim
> KB1RLI
>
>
>