OCPEN ROUTE TO PYPILOT?

[hpeer]

I have OCPN running on an Android phone and a PYPILOT driving a CPT wheel motor. The arrangement works well enough for “point and shoot”. I would like to add the capability to steer to a waypoint and route.

Is this possible?

I believe there was once a plug (created by Sean?) but that it no longer functions?

Is there another method?

[bdbcat]

hpeer...
This should be possible from standard OCPN. But there are some specific details relative to Android, on the hardware side.



As a start, how would you propose to have the Android device communicate with the PyPilot hardware? Wired connection from the Android device? Some sort of WiFi bridge?


Dave

[seandepagnier]

You really only need a tcp nmea connection to pypilot port 20220. Generally you could connect your android device to the same wireless networking to do this. In opencpn you have to also check "output on this port..." which is not by default. This will allow route following without any plugins.

There is the autopilot route plugin which provides alternate calculations for route following overriding the default ones in opencpn.

[hpeer]

Bdcat:

The Pypilot AP provides its own wifi channel and a “control console” to give the AP headings. So the AP and Android are already talking wireless.

Sean:

Thanks for the reply. We were finally able to get underway after nearly 2 years. We did a passage from Delaware to St Thomas. The AP seemed to struggle in GPS mode (drifting off) and was more stable in Compass mode. I still need more familiarity and to test other headings.

Let me try this step by step:
I fire up both pypilot and android.
I connect the Android to the AP wifi.
In OCPN
1- Open CONNECTIONS
2- GENERAL - all boxes UNchecked
3- DATA CONNECTIONS
BOX 1 - ENABLE
BOX 2 - ENABLE.
TYPE - Network
Direction - Input
Protocol -TCP
Network Address - 192.168.4.1 (YDNR connection to receive AIS info)
Network Port - 1456
Priority -1


Now IF this is correct, how do I enable route following?

[seandepagnier]

Basically, you need to add a connection in opencpn to the ip address of the AP on port 20220 tcp with both input and output checked.

It is probably best to continue on the pypilot forum:
https://forum.openmarine.net/forumdisplay.php?fid=18

[hpeer]

Sean;

I appreciate the suggestion to move it there and may eventually.

I would rather handle it here to get input from a different crowd.

That place is probably best for developers and people deeply involved in code.

I need something more for Joe Six Pack, more on my level of comprehension.

Q1: To that end, once I have this connection enabled correctly then I just activate a route?

Q2: Please recall I have a big steel boat. Pypilot compass headings are +100° different from actual or magnetic. Is this a problem? Or does OCPN nust give Port/Stbd correction signals.

The control head is mounted as far from the hull as possible, 6’ off the deck under solar mounts on arch. The compass issue is not going to get better.

[seandepagnier]

Originally Posted by hpeer

Sean;

I appreciate the suggestion to move it there and may eventually.

I would rather handle it here to get input from a different crowd.

That place is probably best for developers and people deeply involved in code.

I need something more for Joe Six Pack, more on my level of comprehension.

Q1: To that end, once I have this connection enabled correctly then I just activate a route?

Yes.

Q2: Please recall I have a big steel boat. Pypilot compass headings are +100° different from actual or magnetic. Is this a problem? Or does OCPN nust give Port/Stbd correction signals.

Did you try entering a heading offset? It does not know which way it faces in the boat depending on how you mounted it.

The control head is mounted as far from the hull as possible, 6’ off the deck under solar mounts on arch. The compass issue is not going to get better.

You mentioned it works under compass mode? Maybe compass is "ok" ?

[hpeer]

Sean,

Thanks for reply.

Yes on the heading off set, as much as 30° to 40°. Compass mode seemed to need significantly less offset than GPs.

The binnacle compass makes a great coat hanger. I don’t have any hope for that. I discovered a magnometer app for the Phone. It has helped located a spot with less interference for the sensor.

As long as the Py Sensor deviation is relatively constant around the entire swing it will be OK. Big jumps would mess it up. Thats why I moved from the CPT sensor, it would work fine on some headings but in other quadrants it could not seem to find a null. BUT the CPT sensor must be within arms reach. I can mount the Py in a much more magnetically friendly location.

We will transit to St Maarten tomorrow I think. I may have a chance to fool with the OCPN route then.

I am also going to try having a Pelagic run my Aeries. The two sensor units are mounted in the same location.

[hpeer]

Sean,

Rereading your reply I may be confused about “heading offset”.

Are you ou talking about Calibration / Magnetic Heading Offset?

Right now it is set to -90.

If I change it nothing seems to change in Control / Heading (while on anchor).

Next, I have set up the connection (in/out) as described above. Entered in a very simple route, and activated the route. Is there any visual indication at anchor that the Android OCPN and PyPilot are communicating?

[hpeer]

Sean (or others)

Question 3: I have set up the OCPN connection as noted including 192.168.14.1/TCPIP/in&out/20220/priority1 hit “Apply” and “OK” and have turned off the instruments connection. Android WIFI set to pypilot, the PYPILOT command screen is active and showing my heading so the Android to PY connection is active. I have a Test route in OCPN and have activated it.
???How do I verify OCPN is talking to the PYpilot and accepting the route???

Question 4: going back to the “offset” issue above. Which “offset”?
A: offset by heading e.g. desire 180 but dial in 200 for a 20° offset?
B: CALIBRATION/Magnetic Heading offset - now set to -90. I fiddle with this but see no change in Heading or Command.
C: CONFIGURATION/Wind Offset - I have not fiddled with this, seems unlikely.

[seandepagnier]

Originally Posted by hpeer

Sean (or others)

Question 3: I have set up the OCPN connection as noted including 192.168.14.1/TCPIP/in&out/20220/priority1 hit “Apply” and “OK” and have turned off the instruments connection. Android WIFI set to pypilot, the PYPILOT command screen is active and showing my heading so the Android to PY connection is active. I have a Test route in OCPN and have activated it.
???How do I verify OCPN is talking to the PYpilot and accepting the route???

pypilot will adjust commanded heading to the active route. Depending on the talker id set, it will automatically switch to gps mode, otherwise you need to manually switch to gps mode.

Question 4: going back to the “offset” issue above. Which “offset”?
A: offset by heading e.g. desire 180 but dial in 200 for a 20° offset?
B: CALIBRATION/Magnetic Heading offset - now set to -90. I fiddle with this but see no change in Heading or Command.
C: CONFIGURATION/Wind Offset - I have not fiddled with this, seems unlikely.

I referred to magnetic heading offset. There is a bug a few versions back that make it impossible to change from the web interface. Try changing it from the ocpn pypilot plugin?

[hpeer]

Sean;

Thanks for the response. One question at a time for the moment.

I will look into the offset thing.

pypilot will adjust commanded heading to the active route. Depending on the talker id set, it will automatically switch to gps mode, otherwise you need to manually switch to gps mode.

This is NOT occurring. Please explain “talker set ID”. Is that to port 20220?

[hpeer]

Sean,

OK that did it, I think, thanks for the help.

IIRC I had tried ti download the PY plugin previously and it didn’t work with my version. Have you updated it or just my poor memory?

In any case I now see the unit finding and adjusting for the route.

Still not sure about the compass adjustment. The plugin has some additional boxes. The adjustment I made was under Calibration/Alignment.

In CALIBRATION/Compass is an untitled box in the lower left corner thT has a slider on the right. It says….
“Insufficient coverage: 6 need 12”
That sounds like satellites, not a compass thing?

[seandepagnier]

it uses measurements similarly to satellites to compute a 2d or 3d compass fix to compensate the sensors.

Dont worry about the heading offset too much as you can always just add/subtract the number displayed in your mind.

At this point, the next thing is to just try with an active route and see if it works.

[hpeer]

Sean,

I believe the PyPilot is correctly steering to the OCPN route. Very good and thanks.

I would like to get a more accurate compass course if possible.

The control head is mounted on the arch, probably not good for pitch/roll but the best option I have for magnetic interference.

Right now the boat is laying about 60⁰ on anchor in Marigot Bay, St. Martin. PyPilot is reporting 270ⁿ. Very nearly backwards. Other headings appeared about 90⁰ off IIRC.

This morning calibration screen shots follow. Does this tell you anything ? Greek to me.