openrcforums

no more than 180 degrees on pan or tilt, regardless of where that circle is with respect to the aerial.

Yes but if the plane had got to 181 degrees for instance the software would have to make the decision to flip the vertical tracking over and then it might decide that it was 179 degrees and flip back, anything on the 180 degrees borderline would cause a continuous vertical flip and horizontal rotation during which tracking would be lost, you need a good margin over 180 degrees to allow for a reliable decision to flip!

What about the decision when the plane is flying near the 360 change point? It might do the full rotating process continuously, right?

Sent from my GT-I9195 using Tapatalk

Yes it looks as though it would.

Guys this is a solved problem. The beam is wide enough for some fuzzy logic/hysteresis.

some fuzzy logic

Mmm..my logic is somewhat fuzzy these days.

jhsa wrote:What about the decision when the plane is flying near the 360 change point? It might do the full rotating process continuously, right?

There isn't a change point as such just sit down and fly your finger around the (180 degree vertical horizontal) array. The so called change point is dealt with by the tilt mechanism.

now bear with me here. I am trying to figure this out. I can understand at the change point you can use fuzzy logic to deal with it.
what I don't get is when the plane is at the other side of the pan180 degree. yes, you can use tilt to the other side and pan it. But what is tracking the pan now? the two pan tracking ant is now facing the other side. If you don't need the 2 pan tracking ant to determine the azimuth angle then why we even need the two ant at all? just use the two telt tracking ant to do the tracking all the time?

You are correct. With an rssi tracker the 180 servo would be a challenge.

Easier with a GPS tracker.

Sent from my GT-I9300 using Tapatalk

kaos wrote:But what is tracking the pan now? the two pan tracking ant is now facing the other side.

Still the pan antennas/servo, its operating direction just needs to be reversed when going past vertical. Remember all antennas move together on a single block.
When you get to the end of the pan range a "snap" requires snapping both the pan and tilt axes.

My idea is to control pan and tilt independently.
Pan, as in Bob's first version, with a continuously rotating servo.
Tilt with a "180 degree" servo. One can then use cheap, strong, metal gear servos.
However, the Act assumes that a wireless transmission of the video signal.

Yes, I have it pinned! If "Tilt" exceeds 90 degrees, then "Pan" must rotate opposite!
Kilrah, I should have read that right - and also understand.

I an at this point seriously considering dropping the tilt.

It just is not needed! Or rather its cool, but does not add much but cost!

It is simpler to just do a 360 rotation, and track pan.

For the central antenna, install a diversity antenna. 2 x helicals. You could effectively get 120 degrees of excellent coverage.

The result is 1 less antenna, 1 less servo and a much simpler sub structure.

I have done both ways, and at this point, for an rssi tracker, I think this is actually the best.



Sent from my GT-I9300 using Tapatalk

I guess the goal and purpose will determine what is needed. with shorter range, things can be simplified a lot. with longer range, things needs to be more tightened up. At this time of age, I think we can make just about any thing we want, but at what cost and how much trouble we want to go.

Kilrah wrote: Still the pan antennas/servo, its operating direction just needs to be reversed when going past vertical. Remember all antennas move together on a single block.
When you get to the end of the pan range a "snap" requires snapping both the pan and tilt axes.

yes, if they are on the same platform. what Rob is having now pan and tilt are separated. 180 degree pan servo just won't do it. 360 degree or continuous servo can, yes.

Using this AAT GPS system (which is simpler to make than RSSI system):
https://www.youtube.com/watch?v=HKCc3vOvqzg#t=23
sold at http://www.foxtechfpv.com/flydream-auto ... p-367.html
for easy description.

if the RSSI ant, diversity .... are all on the white plate, with 2 stepper motor and 2 slipring, it should give the tracker unlimited continuous horizontal tracking without any 'snap back' and better pan resolution compared to a 360 degree servo. Also a 180 to 270 tilt . (making that black box above the slipring a triangular shape to make room for more tilt angle of the ant plate. (the one sold at foxtechfpv only has 90 degree tilt limited by the shape of the black box and hence you still can see some snap back when plane is flying directly across the top).
I looked into the cost of stepper motor. to have the same or slightly larger torque equivalent, the stepper motor and the driver would cost just about the same as the servo Rob is using ~50.00 (the Hitec brand). but I think it would last longer than a servo???

Really like to see the performance comparison in the real between RSSI system and the GPS system. to make the RSSI system, it is going to be big because all the tracking ant are needed. GPS would be much smaller without the need for a big plate for all the RSSI tracking ant.

GPS is much lighter on the ground side, cheaper, simpler, more precise, no hunting,...
The only thing IMO that the RSSI solution has for it is that it doesn't require extra gear onboard.

There were a couple of commercial RSSI solutions at some point, but they were quickly dropped.

I just remembered that FrSky has GPS telemetry. So no extra expensive downlink required really.
If the downlink is relatively feast and the coordinates precise.

Downside of the frisky gps is a link from the radio to the tracker. You have yet another wire!

What would be great is a standalone 'sniffer' that could listen in on the link, and feed the gps data.

Sent from my GT-I9300 using Tapatalk

that is why I was excited when Brentrand was thinking about the GPS tracking function in the Tx. Basically no extra onboard equipment beside a GPS which the multi rotor or ardupilot /megapirate NG already has.

I am particularly interested in Bertrand's idea of handling the tracking within radio Tx. Imaging the Frysky GPS downlink to Tx and the Tx will calculate the tracking and send the tracker servo commend via the 2nd tx module to the tracker with a second frysky receiver on on the tracker. I hope Bertrand will keep pursuing that subject.

The AAT system actually use audio ch to transmit GPS data to ground station via vTx.

Ah.... So a secondary RX in then tracker which receives the signals to move to x location?

That would be clever!

Maybe it will be possible with lua.

Sent from my GT-I9300 using Tapatalk

Yep, that's the idea.
Interested in getting the code written for 9X? YOu have not written some 9X code for a while right!

Certainly will give it some thought!

think hard! coffee, beer is on me.

Rob:
Here is what I have in mind of the the tracker: I think it works both for GPS or RSSI. just put Rssi ant on the panel.
This is based on the Foxtech AAT system with a little mod, two stepper motor and two slip rings. The AAT 12 ch version has one 12 ch slip ring on the bottom. This will give 360 degree horizontal and 270 degree vertical continuous rotation/tracking.

I am surprised that no one has looked at this -----> Pepperbox Phased Array

Interesting. in the spec it says 145 degree while IB crazy video he says 120 degree. Never the less that is a 'wide' coverage and long. That would make tracking a lot easier and not requiring much precision. I wonder if you can use it as a transmitter ant? IB crazy mentioned using other ant as Tx and this as Rx ant. Could use this as radio Tx ant??

The antenna isn't worried if it is used for TX or/and RX.

I wonder if I can put a video rx helical ant in the middle of that box so radio tx and vedio receiving can be tracked at the same time (using GPS method)? Any radio interference issue?

It will be fine.



Sent from my GT-I9300 using Tapatalk

Great. How is your RSSI system coming along? I still yet to see how well the RSSI system compared to GPS system.

Lol..

Check rcgroups. Lots of similar systems.

I have actually diverted my attention to a GPS option.

Can't say too much beyond the fact it will use some clever frisky magic to handle the GPS tracking.

Sent from my GT-I9300 using Tapatalk

Just math.
got a link to rcgroup of the good ones? Too many of those for me to filter out.

This look an quite good.

http://www.rcgroups.com/forums/showthre ... p=26471708



Sent from my GT-I9300 using Tapatalk