Welcome, Guest. Please Login or Register

Author Topic: Locally developed Rally Navigation Eq.  (Read 14399 times)

0 Members and 1 Guest are viewing this topic.

Offline SteveD

Re: Locally developed Rally Navigation Eq.
« Reply #20 on: January 20, 2015, 05:12:50 pm »
Sub, extremely interested  :sip:
 

Offline Weedkiller - Adie

Re: Locally developed Rally Navigation Eq.
« Reply #21 on: January 20, 2015, 05:54:19 pm »
** Snip **
ico-rallye-cap, which doesn't use the wheel sensor at all, but gets the GPS  to provide the distance.  This removes the potential failure point of a cable, a sensor and a magnet.  The installation becomes instantly more user friendly to install and more compact.
** Snap **

As per above we got no choice but to go with a GPS. My original design used the GPS to do 'everything' but then we opted for the Mag Compass. Now we need both Mag Comp and GPS.  ???  Fortunately it will actually make the programming easier with only a GPS and buttons.

** Snip **
As someone with super-bright LEDs on his bike at the moment, I can testify that they don't work too well if they're in direct sunlight.
Do you think the GPS could be integrated into the unit (sit below the LCD? or on top of it, angled forwards etc)? .
** Snap **

If you look at most other Trip meters (ICO as well) the basic distance/speed display need some sort of 'indicator' so the user is informed what they are looking at.  As we start with a clean slate we will try to make the system easier to use but still be 'compatible' with 'tradition'.  :o

Size wise we can fit the GPS unit inside the unit.  We already checked that and also think that will be a moerse bonus.  Not sure about the electronics above (display) and below (processor) affecting the areal.

Dankie vir al die insette, laat dit kom, ek moet teen die naweek toetse begin doen.!!


Adie
Laat ons Stof maak.
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #22 on: January 20, 2015, 05:58:36 pm »
** SNIP **
Or, you could just add(for extra cost) a RS232 to TTL converter and read the NMEA sentences from the GPS directly on the micro controller, use the coordinate at the point to work out the deviation and use that constant as offset on the z-axis compass you are using.
** SNAP **

This got me thinking  :laughing4:
1. GPS  A GPS does not seems to have a Compass. It requires the unit to move in order to convert the co-oords of the two positions into the 'Bearing' (Tested the theory with a Garmin unit.)

** Start **
$GPRMC,161229.487,A,3723.2475,N,12158.3416,W,0.13,309.62,120598,,,A*10
Message ID $GPRMC  RMC protocol header
UTC Time 161229.487  hhmmss.sss
Status   A    A=data valid or V=data not valid
Latitude  3723.2475  ddmm.mmmm
Longitude 12158.3416  dddmm.mmmm
Speed Over Ground 0.13 knots  
Course Over Ground  309.62 degrees  True  <---------------**
** end **

2. A Cellphone got a Magnetic Compass AND a GPS.  It can thus give both headings.

NOW I remember. I chose the Mag Compass because it will give the heading even when stationary whereas the GPS must move at least 3meters or pick up different co-oords to do the 'Course' and then calc the True North.

In order to get FULL functionality We will need both the Mag compass (Stationary heading) and the GPS (True North when moving)  Fortunately it will only require 2 lines of code to select the heading type between moving and stationary.  The software already reads both devices.

this is getting interesting.  Keep it comming.

Adie





You do not have to have the GPS moving and use the vector to calculate the bearing.  You only need the coordinate to calculate the offset from Magnetic North.

The intentions on my unit was to have a configurable offset; this is obviously not sufficient in most cases, but if you know you going north/south with little deviation on the longitude it could be sufficient.  In rallye the compass bearings really have a tolerance of (wild guess) 15%(although, the Amageza it was pretty damn spot on!).

I did however research(and wrote code) to use NMEA sentences and use it to offset my bearing from the real compass.  You often find yourself completely stationary, missing a turnoff of some sorts and referencing your CAP.  Your turn your handlebars or bike slightly, or spin it around and get the correct bearing(if you have an electronic magnetometer) - with GPS, your have to effectively travel about 20 meters for it to recalc, since 1. you need a good vector, 2. your software has some averaging(hopefully), prohibiting it from giving your a bearing definitively.

On "bearing"/CAP, you have to make very sure that you do not print out every reading from your magnetometer.  They are normally quite quick with the updates and you will have crazy degree jumps!  If you naively build in averaging by adding numbers in an array, you will get the problem where, if you lets for example say measure 4 readings for your average : 2 of them is 5deg, and 10deg, the other 2 is 350deg, 359deg - your average will be 180deg!!! which is wrong!  Therefore, and I am possibly teaching my grandma to knit here, you should use the mean angle of a array of cirular quantities(degrees/radians).  The formula for this was proven by a few and in different ways.

So, a few things to consider...
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #23 on: January 20, 2015, 06:05:26 pm »
Size wise we can fit the GPS unit inside the unit.  We already checked that and also think that will be a moerse bonus.  Not sure about the electronics above (display) and below (processor) affecting the areal.

This is possible, but a very dangerous solutions.  GPS antennae like to be pointed in there designed direction (receiving angle[cone] upwards).  That is easy, either you put the whole unit as a separate wire-connected unit on top of the nav tower or you have a remote antennae that you can point better.

Then, on GPS!  Hobbyist SiRF GPS chipsets generally takes a very long time to acquire a fix, from 30 secs to 2 minutes(sometimes you are lucky and hit 15 secs)  SO :) What you need to consider is a external battery on the GPS board/section, i.e. just fckoff big electrolytic capacitors.  That means the GPS will hold it's fix for a minute perhaps, even if power is cut.   You cannot drive the GPS and the microcontroller on pill 2032 batteries, it will drain them in no time.
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #24 on: January 20, 2015, 06:05:58 pm »
I will stop now  :peepwall:
 

Offline SteveD

Re: Locally developed Rally Navigation Eq.
« Reply #25 on: January 21, 2015, 08:26:24 pm »
Don't stop, what you saying is all true.
Have a look at https://tiny-circuits.com/tiny-shield-gps.html
It's the smallest integrated GPS board that I have seen yet, and they claim pretty good sensitivity and time to first fix.
What is your market? Dakar will not allow anything with a GPS in it, besides the ERTF SPOC for some reason. They specifically did not allow the ICO Rally CAP. If you stick the GPS inside, you can't sell it to people riding Dakar.
Another alternative would be to use a larger external GPS "puck". Garmin are offering one again, the GPS-18.   http://www.garmin.co.za/prod_detail.php?productid=264&categoryid=&subcategoryid=

Regarding your power supply:
You will see very large surges on the 12V input. Not very often, only during load dump events, but one of those can smoke a supply quite spectacularly. I would cater for input surges of at least 65V, preferably 87V. ISO 7637-2 is the best guideline that have found yet, and it's free. Test pulse 5a is the one that does the damage. I would design preferably to operate during load dump, alternately survive the load dump and operate afterwards.

Now I'll keep quiet for a while......
 

Offline Weedkiller - Adie

Re: Locally developed Rally Navigation Eq.
« Reply #26 on: January 21, 2015, 10:11:05 pm »
PLEASE DONT STOP.

Most of what you both said are in my design specification.  Not in detail but things I had to consider.

@DV, I use a fast filter to 'smooth' the 'jittery' of the compass if the CAP is mounted on the handlebars (User settable)  similar to those used in some loadcell amplifiers to 'stabilize' the reading on the scale. This also take the 360/0 changeover into consideration.  With respect to compass, the system will sense the bike is stationary and select the magnetic as primary.  If I wanna be fancy we can correct the bearing based on the GPS co-oords (will probably do that) and then use the GPS 'course' value once moving.  The nice thing with this method is that we only need to do the MN/TN calc once and save computation time. (the bike is standing so not much happens in any case  :) )  This will ensure that the compass is 'active' even if stationary. I fried my GPS module so wait for a new one. :-[

@ SteveD, Thanks, I do currently protect 'permanently' from 24Volts (run a 12V and then 5V regulator) because of the Dakkies known over voltage from the regulator.  Will look at 'powerline protection' if I understand it correctly. (I am mechanical and only know electrics can shock.!!!

With respect to the built in GPS, will do some research. I think the problem with a GPS was that in 2007?? they tried the ARROW heading but think it was then too easy as you could probably set the co-oords of the waypoint and it will be to easy NOT to miss. (Just thinking!)

How do the Dakar riders get True north if no CAP heading instrument?

Dont stop, keep it coming.

Adie
Laat ons Stof maak.
 

Offline Bill the Bong

  • Coffee Monster
  • Race Dog
  • ***
  • Bike: Cagiva Elefant 900
    Location: Northern Cape
  • Posts: 4,921
  • Thanked: 246 times
Re: Locally developed Rally Navigation Eq.
« Reply #27 on: January 21, 2015, 10:23:56 pm »
They use a cap repeater from the ERTF
 

Offline Bill the Bong

  • Coffee Monster
  • Race Dog
  • ***
  • Bike: Cagiva Elefant 900
    Location: Northern Cape
  • Posts: 4,921
  • Thanked: 246 times
Re: Locally developed Rally Navigation Eq.
« Reply #28 on: January 21, 2015, 10:34:22 pm »


With respect to the built in GPS, will do some research. I think the problem with a GPS was that in 2007?? they tried the ARROW heading but think it was then too easy as you could probably set the co-oords of the waypoint and it will be to easy NOT to miss. (Just thinking!)

How do the Dakar riders get True north if no CAP heading instrument?

Dont stop, keep it coming.

Adie

2 things:  there's not nearly enough waypoints to use as a primary navigation means.  The next waypoint may be 15km away on the other side of a mountain.  The other thing is that if you do try and nav WP to WP, you will miss tulips.  Every tulip counts for penalties as a WP, even if there's no published LATs / LONGs.
 

Offline Weedkiller - Adie

Re: Locally developed Rally Navigation Eq.
« Reply #29 on: January 21, 2015, 10:56:14 pm »
OOPS. It was 2005 with the arrow and not 2007.
** snip **
In my opinion, the reason they donít have fancy devices telling you exactly which way you should always be heading is because they want to keep it difficult. Itís too easy to just follow an arrow.

In Dakar 2005 (only), they used the arrow approach, where the GPS always told them which direction they should go. In 2006 they went back to the hard way.
** Snap **

@Bill,
** snip  from 2007 Dakar docs. **
GPS in the Dakar:
In the Dakar, the GPS has very limited function.  The rules change somewhat each year,
so check the rule book.  Generally, the GPS will give you only your CAP heading.  There is
usually no arrow pointing to a waypoint ahead.  So, you navigate using three  sources of
information:  ICO for kilometers, GPS for CAP heading, and Roadbook for instructions.  
 
In certain circumstances, the GPS will show an arrow directing you to the next waypoint.  
This is usually only within a prescribed radius of the waypoint.  
** Snap**

I was merely wondering how the Dakar riders get the CAP without a (their own) GPS based on Steve's post.

Adie

« Last Edit: January 21, 2015, 10:58:59 pm by Weedkiller - Adie »
Laat ons Stof maak.
 

Offline SteveD

Re: Locally developed Rally Navigation Eq.
« Reply #30 on: January 21, 2015, 11:55:24 pm »
I was merely wondering how the Dakar riders get the CAP without a (their own) GPS

Dakar do not allow you to have your own GPS, they mandate that you use the ERTF unit. You may use an approved CAP repeater, but it has to be fed from the ERTF unit  :thumleft:
 

Offline alanB

  • Moderator
  • Grey hound
  • ****
  • Bike: Husqvarna (all models)
    Location: Gauteng
  • Posts: 6,767
  • Thanked: 11 times
    • www.extinction.co.za
Re: Locally developed Rally Navigation Eq.
« Reply #31 on: January 22, 2015, 06:20:58 am »
Great work guys keep it up  :thumleft:

Fascinating thread!
Husqvarna '09 610TE - Great Bike!

I just finished a SciFi novel Extinction: Task Team, download the preview.
 

Offline Weedkiller - Adie

Re: Locally developed Rally Navigation Eq.
« Reply #32 on: January 22, 2015, 07:07:24 am »
** Snip **
Dakar do not allow you to have your own GPS, they mandate that you use the ERTF unit. You may use an approved CAP repeater, but it has to be fed from the ERTF unit.
** Snap **

@SteveD, AHHHH, that make sense.

I will look at maybe make the GPS to fit 'underneath' the CAP unit instead if inside and still retain the install simplicity and marketability.  Most units are 10mm thick.  In my search for GPS units we definitely consider external units as they are fairly common and the connection to the Atmega/PIC is not a problem.

@alanB
Even if we fail to produce a marketable product the information passed around do have value for rally raiders. (I hope)

Adie
Laat ons Stof maak.
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #33 on: January 22, 2015, 07:21:32 am »
Sorry for the hijack Adie, but this seems like a general homegrown RallyeNav thread.

I think I have decided, and this might go back on what I have said before, and that is that I am going to drop the compass idea for my next version.  I am going to calculate bearings using my moving vector and display that. 

The spec also allows, without calculation, to display a bearing to the next waypoint.  This way I have a ERTF/CAP repeater in one!

I will still test drive my current unit, but I am afraid that there is too much shit in it already :)

 

Offline alanB

  • Moderator
  • Grey hound
  • ****
  • Bike: Husqvarna (all models)
    Location: Gauteng
  • Posts: 6,767
  • Thanked: 11 times
    • www.extinction.co.za
Re: Locally developed Rally Navigation Eq.
« Reply #34 on: January 22, 2015, 07:28:49 am »
What are your thoughts (Garth, Addie, et al) on the Rally Blitz app on an iPhone?

I was really surprised on how well this seemed to work on the last Amageza.

The App is cheap and many people have iPhone's so that represents potentially a very cost effective approach for many people?
Husqvarna '09 610TE - Great Bike!

I just finished a SciFi novel Extinction: Task Team, download the preview.
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #35 on: January 22, 2015, 07:34:25 am »
I could not see shit on an iPhone, some of the guys had it mounted in the "shade" part of their nav tower and that seemed to work.  As Kamanya, he had great difficulty reading his iPhone in the sun.
 

Offline alanB

  • Moderator
  • Grey hound
  • ****
  • Bike: Husqvarna (all models)
    Location: Gauteng
  • Posts: 6,767
  • Thanked: 11 times
    • www.extinction.co.za
Re: Locally developed Rally Navigation Eq.
« Reply #36 on: January 22, 2015, 07:45:35 am »
I could not see shit on an iPhone, some of the guys had it mounted in the "shade" part of their nav tower and that seemed to work.  As Kamanya, he had great difficulty reading his iPhone in the sun.

Thanks  :thumleft:

Would you say its fine as long as its in the shade, or would you say its still a problem but doable in the shade?
Husqvarna '09 610TE - Great Bike!

I just finished a SciFi novel Extinction: Task Team, download the preview.
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #37 on: January 22, 2015, 07:46:43 am »
I think it is doable, but not ideal.  I think MaxThePanda had an iPhone as well? Perhaps give him a shout.
 

Offline mtbbiker

  • Race Dog
  • ***
  • Bike: Suzuki DR-Z 400
    Location: Gauteng
  • Posts: 515
  • Meerkat V2.0
Re: Locally developed Rally Navigation Eq.
« Reply #38 on: January 22, 2015, 08:17:45 am »
Sorry for the hijack Adie, but this seems like a general homegrown RallyeNav thread.

I think I have decided, and this might go back on what I have said before, and that is that I am going to drop the compass idea for my next version.  I am going to calculate bearings using my moving vector and display that. 

The spec also allows, without calculation, to display a bearing to the next waypoint.  This way I have a ERTF/CAP repeater in one!

I will still test drive my current unit, but I am afraid that there is too much shit in it already :)


Sorry for my ignorance, but how does the ERTF/CAP Repeater work, do you upload the way point coordinates to the device before the race or do you have to "enter" the value when reaching a way point. I have played with that feature on the Rally Blitz app (although just riding the mtb bike) and It works great when doing an unknown route.
Without goals and plans to reach them, you are like a ship that has set sail with no destination - Fitzhugh Dodson
Project Meerkat:DRZ rally build http://www.wilddog.za.net/forum/index.php?topic=86025.0
 

Offline darthvader

Re: Locally developed Rally Navigation Eq.
« Reply #39 on: January 22, 2015, 08:19:03 am »
Sorry for my ignorance, but how does the ERTF/CAP Repeater work, do you upload the way point coordinates to the device before the race or do you have to "enter" the value when reaching a way point. I have played with that feature on the Rally Blitz app (although just riding the mtb bike) and It works great when doing an unknown route.

It is uploaded for you I think.