• 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
First test, successfull!
#1
Hi,

Last Saturday I did a first trial. Just a quick one, with a lot of tape and wires all through the boat.
   
   
   
   
   

(07-30-2018, 12:54 PM)It Paradyske Wrote: Hi,

Last Saturday I did a first trial. Just a quick one, with a lot of tape and wires all through the boat.

And the first try:

The rudder went in the corner and stayed there.

Of course that was something we didn't expect, because I thought I connected it in the right way.
Well actually I was not surprised..

So I reversed polarity of the motor, and tried again.

And yes!!!!
It worked!!!
For a short while, than the motor controller stopped.
Don't know why so I resetted it and tried again, and it stayed working!!
   
It was only a very short test trip, but I am confident enough to continue now :-)

Two things that came up in my mind:

Wouldn't it be useful to have a polarity switch somewhere in the calibration window?
Perhaps a save button for different gain settings, so you can switch between different settings easily?
(Storm, no wind and something in between?)

So some things to sort out:
why my first two IMU's behave so badly
why the motor controller needed a reset

how do I get the GPS and wind options enabled, I can only use compass.
(The openplotter configuration had a GPS, and I had activated a waypoint)

Thank you Sean!
  Reply
#2
that looks promising, congratulations! thanks for reporting
  Reply
#3
Quote:Wouldn't it be useful to have a polarity switch somewhere in the calibration window?
I knew this would be asked

Now, the autopilot should maybe automatically flip the polarity, but this logic is not ready.

Having a polarity switch is extra gui clutter when all you really need to do is swap the wires.

If you wanted to be able to use the autopilot motoring backwards it would be useful.
Quote:Perhaps a save button for different gain settings, so you can switch between different settings easily?
(Storm, no wind and something in between?)
Also a great idea, however, it is not so simple, because ideally the gain settings would be stored on the server, not in the opencpn plugin so they would be available to any autopilot control client.

Another option would be to re-formulate the gains so that there are only one or two main ones that are changed for different conditions.
Quote:So some things to sort out:
why my first two IMU's behave so badly
Not sure what type of imu you had, but many require accelerometer calibration.

I had been using mpu9255 all along which "usually" do not require this.. some do, and most other sensors do. I just added accelerometer autocalibration a few days ago, and it has a special procedure.
Quote:why the motor controller needed a reset
It's difficult to say since this is your homemade motor controller. My controllers all automatically reset if no data is received, There are many possible reasons, and even depending on the fuse bytes in the arduino. You would need to reproduce this and debug it. My motor controllers do not do this.


Quote:how do I get the GPS and wind options enabled, I can only use compass.
(The openplotter configuration had a GPS, and I had activated a waypoint)

Thank you Sean!
To get gps data you can run gpsd, and assign your gps to it. pypilot automatically uses gpsd, and opencpn can easily add this connection.

You can assign serial ports to pypilot. If they have wind or gps, they will also be relayed over tcp to other programs like opencpn on port 20220. This is ideal because the autopilot gets the data with minimum latency.

You can also send the nmea data to pypilot via port 20220. So if opencpn has wind/gps already, then make a connection to tcp port 20220, and make sure it is marked for output. This will add latency to the measurement, but still works in practice.
  Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)