Plugin: DashBoard

[rgleason]

Mathew Durst Post #361

The real interesting thing, though, is that the MPU-9150 also calculates rate of turn/pitch/roll, which I also ran through the Arduino to correct for the induced wind at the top of the mast from how fast it's pitching/rolling/turning.

rate of turn/pitch/roll that's acceleration! If you could use that effectively it might really help.

The nice thing about the MPU-9150 is that the code I'm using includes a simple Kalman filter. I certainly wouldn't use it as an Inertial Navigation Unit (such as you see in some older jets before GPS), but it holds it's accuracy very well so long as you don't really move for the first five seconds. I suppose if you reset it at sea, then it probably wouldn't work to well.

I'm also trying to rework the code so that it gets at least a 10hz resolution, but the problem I'm running into is the sentences start writing over each other when it runs that fast. Another problem is that I'm not a programmer, so it's a slow process.

Mathew, I wish you success in building a 10mhz with good accuracy!
King Tide Sailing
How to Setup the MPU-9150 9-Axis Accelerometer, Gyro, & Compass with an Arduino
Correcting NMEA-0183 Wind for Vessel Roll, Pitch, and Yaw
How to Connect any NMEA-0183 Device to an Arduino
Set and Drift; or, the Speed and Direction of the Tidal Current (NMEA-0183 VDR)
An Arduino NMEA-0183 Multiplexer
How To Install OpenCPN on a Raspberry Pi

I have an old Raymarine Fluxgate outputting nmea 0183 to the Raymarine ST400 Wheelpilot which does not seem to move heading data onto the Seatalk bus. Therefore I should probably be making or buying a new heading sensor that is more accurate. Will this one work well enough for radar overlay? I am probably going to need a multiplexer too.

[TBillings]

Originally Posted by rgleason

King Tide Sailing
How to Setup the MPU-9150 9-Axis Accelerometer, Gyro, & Compass with an Arduino
Correcting NMEA-0183 Wind for Vessel Roll, Pitch, and Yaw
How to Connect any NMEA-0183 Device to an Arduino
Set and Drift; or, the Speed and Direction of the Tidal Current (NMEA-0183 VDR)
An Arduino NMEA-0183 Multiplexer
How To Install OpenCPN on a Raspberry Pi

I have an old Raymarine Fluxgate outputting nmea 0183 to the Raymarine ST400 Wheelpilot which does not seem to move heading data onto the Seatalk bus. Therefore I should probably be making or buying a new heading sensor that is more accurate. Will this one work well enough for radar overlay? I am probably going to need a multiplexer too.

It should, but depending on what sort of NMEA sentence is used, you may have to tweak the code a bit (which would be very, very easy to do).

If all you need is a compass, then you can easily do 10hz resolution. I did some ops testing this weekend for a new method of writing the sentences, but I still ran into the same problem--the GPS and Transducer Sentences read too slow, and get stepped on by the IMU (the gyro/accelerometer/compass). In fact, I had to reduce the resolution down to 1hz because OpenCPN kept dropping the GPS signal. Of course, that was under the "max performance" setup, where it output the following sentences:

- RMC, GPS info (lat/long/speed/course/etc...)
- DPT, Depth (corrected for transducer height and offset angle from vertical)
- MTW, Mean temperature of the water
- VHW, Vessel Hull Speed through the Water
- HDM, Magnetic heading
- HDT, True Heading
- ROT, Rate of Turn
- XDR, Pitch/Roll/Rate of Pitch/Rate of Roll (emulating the NKE devices, but for $40!)
- VDR, Set and Drift (calculated from all the inputs)

Of course... Dashboard doesn't seem to incorporate ALL of these, and certainly you don't need ALL of these sentences... Realistically, in practice I only use the RMC, HDT, and VHW sentences (maybe VDR if I can figure out how to incorporate it in OpenCPN). But this past weekend I was trying to, like I said, max perform the device. And it appears as though if you want to multiplex ALL those sentences in one Arduino, the fastest resolution is 1hz.

However, if you only want the IMU sentences (Headings, Rate of Turn, Pitch/Roll, Accelerations), then you can definitely and easily use 10 hz (and probably faster too). I can't imagine any recreational application that would need more than 10hz, including anything for the radar. Rasbats seems to have made one for his radar, so I'd ask him for pointers. As far as accuracy goes, that depends on a few things. One, you MUST mount the compass in a way that truly points forward for your vessel. Usually a bulkhead is straight up and down and left and right, so you could mount it there. I have mine flush against a bulkhead inside my Nav desk. Two, you MUST calibrate the compass using the program supplied on my blog. I go into more detail there, but the basic reason is that it stores the calibration data on the Arduino itself--most calibrations require a third party program on a windows computer, which I'm assuming most people don't want to include on the boat. Also, keep that iron and magnetic material as far away as possible. And try to keep it still for at least 5-10 seconds after power application.

After all that, I've found it to be quite accurate (within 5 degrees in my slip). But like I said before, I wouldn't use it for any commercial applications because this is a DIY project.

Also, keep in mind, this is a one way gate; it only outputs data, it can't do two-way comms on the Arduino. Now, if you follow my guide and incorporate the Raspberry Pi as a WiFi multiplexer and use Stripey Dog's Kplex, then you could multiplex it all (my next step is to figure out how to output OpenCPN autopilot data through the RPI and into my autopilot).

Another alternative I suppose is to use an Arduino Uno (the cheap one) for the IMU, and then use the Mega for all the other instruments. That way you could multiplex everything as fast as possible... which actually isn't that bad of an idea!

[rgleason]

Mathew Thank you for the guidance and detail. Very helpful

[shallowain]

hey everyone, does some one know if there's plans to implement the use of MWT sentences for True Wind Direction and Speed in the Dashboard?
my sensor unfortunately does not output VWT

[TBillings]

Originally Posted by tom_BigSpeedy

Rick,

don't take it too serious with the current diplay (and naming) of the performance data in the "text style instruments", please. I gouped the stuff together for now to verify the data against the loaded polar.
* percentage values are always the reference from your momentary speed
to the calculated (optimum) speed. It's simply STW/<opt.speed>*100
* The brown (orange) arrow in the compass instrument is AWA
* Blue arrow is TWA, correct.
* Polarspeed is the speed your boat can theoretically do based on the
polar data at the current wind angle and speed. Useful if you cannot go for the target-VMG angle for whatever reason, or if you're reaching instead of going up-/downwind.
* Performance VMG = actual VMG, correct. I will rename it to actual VMG.
I called it Perf.VMG as there is already a "VMG" instrument in dashboard.
* the blue centered arrow (in the compass instrument and in the chart) is the (sea) current.
* Target VMG = target boatspeed at target VMG (+% value as reference to your momentary STW)
The vertical layline in the compass instrument points to COG, so it may move aside, the width of the laylines represent your boat yawing around the vertical axis (needs still some improvement), so you really see the "target/waypoint area" you're COG indicator points to.
There's a right click menu on the dashboard to show/hide the laylines in the chart.
There's a checkbox in the dashboard settings to show/hide the current display (blue arrow) in the chart.

I can already place a temporary waypoint on the chart from the plugin.
Goal is -- as in Baikals screenshot -- to print the target-VMG angles onto that Waypoint and intersect it with your laylines. But that's to be done yet.
I hope that clearifies it a bit...

Thomas

Any update on this Tom?

[rgleason]

implement the use of MWT sentences for True Wind Direction and Speed in the Dashboard?
my sensor unfortunately does not output VWT

This is a perfect one for Dirk's Nmea converter. Look at the docs. Tell us how you used it in detail. I will post that in the wiki for others.

[rgleason]

Haven't heard from big tom, but sure would like to use his plugin.

[arnolddemaa]

Hey,

Can anyone tell me how the barometer function must be controlled?
I tested using the NMEA converter
MDA, MMB and XDR. But unfortunately without success.
So the question is to what NMEA code listens barometric


ps my B&G H3000 sent
$IIXDR, C, 33.9 C, Airtemp, P, 0998, B, BARO, A, -34.0, D RUDDER * 04

thanks,
Arnold

[conachair]

Originally Posted by arnolddemaa

Hey,

Can anyone tell me how the barometer function must be controlled?
I tested using the NMEA converter
MDA, MMB and XDR. But unfortunately without success.
So the question is to what NMEA code listens barometric


ps my B&G H3000 sent
$IIXDR, C, 33.9 C, Airtemp, P, 0998, B, BARO, A, -34.0, D RUDDER * 04

thanks,
Arnold

Try the barometer history with the MDA sentence,see here...

http://www.cruisersforum.com/forums/...ml#post2037645

And a bit more here...
New Plugin NmeaConvertor - Page 2 - Cruisers & Sailing Forums

[arnolddemaa]

Originally Posted by conachair

Try the barometer history with the MDA sentence,see here...

http://www.cruisersforum.com/forums/...ml#post2037645

And a bit more here...
New Plugin NmeaConvertor - Page 2 - Cruisers & Sailing Forums

Sorry it does not work. I am tested with:
$WIMDA,30,I,1.02,B,10.1,C,,C,,,,C,158.69,T,,M,3.89 ,N,2,M*37
and the new nmea converter the new with the decimal places

Tested on windows 10 with OpenCpn 4.2

What I am doing wrong?

Arnold

[conachair]

Originally Posted by arnolddemaa

Sorry it does not work. I am tested with:
$WIMDA,30,I,1.02,B,10.1,C,,C,,,,C,158.69,T,,M,3.89 ,N,2,M*37
and the new nmea converter the new with the decimal places

Tested on windows 10 with OpenCpn 4.2

What I am doing wrong?

Arnold

Try
$WIMDA,30,I,1.02,B,10.1,C,,C,,,,C,158.69,T,,M,3.89 ,N,2,M
Without the *37 at the end.

Sent from my SM-G357FZ using Tapatalk

[transmitterdan]

In O 4.2 the numeric barometer instrument is broken. The history works. Will be fixed in a soon to be released update I believe.

[arnolddemaa]

Originally Posted by transmitterdan

In O 4.2 the numeric barometer instrument is broken. The history works. Will be fixed in a soon to be released update I believe.

Thanks. I will test on o 4.0

Arnold

[rgleason]

The VMG dial does not seem to work. SOG is there, but the VMG is not calculated.
What data is necessary? Is this VMG to mark or VMG to windward or something else?


Dashboard_pi.cpp line 2389
[QUOTE]case ID_DBP_I_VMG:
instrument = new DashboardInstrument_Single( this, wxID_ANY,
getInstrumentCaption( id ), OCPN_DBP_STC_VMG, _T("%5.2f") );
break;
case ID_DBP_D_VMG:
instrument = new DashboardInstrument_Speedometer( this, wxID_ANY,
getInstrumentCaption( id ), OCPN_DBP_STC_VMG, 0, g_iDashSpeedMax );
( (DashboardInstrument_Dial *) instrument )->SetOptionLabel( 1,
DIAL_LABEL_HORIZONTAL );
( (DashboardInstrument_Dial *) instrument )->SetOptionMarker( 0.5,
DIAL_MARKER_SIMPLE, 2 );
( (DashboardInstrument_Dial *) instrument )->SetOptionExtraValue(
OCPN_DBP_STC_SOG, _T("SOG\n%.2f"), DIAL_POSITION_BOTTOMLEFT );
break;[/QUOTE]

[rgleason]

I can't really understand the code below, but awhile back I wrote this:

There are two types of Calculated VMG Velocity Made Good that would be very useful, in addition to VMG from instruments.
1. VMGm = Calculated VMG to Mark
2. VMGw = Calculated VMG to Wind Direction
3. VMGn = From Nmea and instruments
Perhaps these should all be separate items in Dashboard?

Why?

When sailing upwind or downwind, VMGw (to Wind direction) is a much more useful gauge because it remains constant and is a good yardstick for maintaining optimum angle and speed.

When sailing upwind or downwind, VMGm (to Mark) varies the closer you get to the mark.

Would it be possible to impliment this please? I've put in Tracker.