Replies to This Discussion

ModeratorPermalink Reply by Chris Anderson on January 3, 2011 at 6:26pm

With the current code, you need to do a warm reset of the APM board to configure the MediaTek properly (this is in the manual). We'll be releasing a version of the code that fixes that tonight.

 

No idea what's going on with your GCS. It's rock solid for me.  

• ▶ Reply

Permalink Reply by Skip Frombach on January 3, 2011 at 7:24pm

Hi Chris,

            I did the warm reset after I saw that mentioned on the forum posts.  No difference for me although I do see the message on the com port window of the Happy Killmore GCS say that a message of 103% has been sent.

Nothing happens with the GPS lock although I can see Longitude and Latitude displays on the GCS.  I also played around with the sampling rates for the altitude, waypoint and location to see if that would change things.  Also sent data at every rate on the com port and those listed in the code.  No change.  I still can't get an actual "GPS Lock" indication.  The MediTek upgrade seemed to load OK.  I got the green circle on the program when it loaded. 

• ▶ Reply

ModeratorPermalink Reply by Ken on January 3, 2011 at 6:28pm

After getting GPS lock (per the receiver) did you tell the APM to reset?

 

• ▶ Reply

Permalink Reply by Skip Frombach on January 3, 2011 at 7:31pm

I didn't tell the reciever to reset with code.  No, I didn't do that.  I pushed the reset button on the Oilpan IMU.  Is there a line of code that I need to add or something that I am missing here?

The other puzzle is what differs from the original Ardupilot Ground Station code that is not the same on the Legacy version that would cause the violent reaction of Google Earth on the Happy Killmore GCS.  That is just wild.  If you click on any of the view buttons it instantly shoots you off the planet and 70,000 miles out into space.  It also inactivates the scaling buttons on Google Earth to where they swing around and then act like a pendulum. 

I may have to write Happy Killmore and see if he has any ideas on what could be happening.  I am just surprised that no one else is having a similar problem.

• ▶ Reply

ModeratorPermalink Reply by Chris Anderson on January 3, 2011 at 8:50pm

Skip: there is no "GPS lock indication" in the GCS with the current APM code. If you're seeing Lat and Lon in the GCS everything is working fine.

 

I have no idea what's going wrong with your Google Earth. I've never seen that. What version of HK GCS are you using?

• ▶ Reply

Permalink Reply by Dale Everett on January 27, 2011 at 11:09pm

I'm having the same issue with HK's GCS where even when I get lat and long values that are correct. Distance above the earth is usually just around 5000 miles. Using latest versions of both APM and the GCS. Will keep poking around.

• ▶ Reply

Permalink Reply by Victor on March 9, 2011 at 8:56pm

Is this a problem with APM 1.0 or with all the APMs?

 

I don't think the problem is related with Google Earth or HK GCS because if  when I test my GPS using the CLI in test mode it says #sats: 0 just like in the picture http://code.google.com/p/ardupilot-mega/wiki/GPSInput.

 

Is this a Mediatek problem? APM coding? APM board? what about ublox?

• ▶ Reply

ModeratorPermalink Reply by Chris Anderson on March 9, 2011 at 9:00pm

Victor: I'll repeat what I've now said here countless times. There is no "problem". In APM 1.02 we just don't pass along that data. The autopilot has the data and uses it, but doesn't resend in the telemetry stream. 

 

By popular demand, we've now added it to APM 2.0, which we'll release this week. No biggie.

• ▶ Reply

Permalink Reply by Victor on March 10, 2011 at 4:48pm

Thanks Chris for the clarification. I'll look foward to that soon. How much longer will diydrones be giving support for APM 1.02?

• ▶ Reply

ModeratorPermalink Reply by Chris Anderson on March 10, 2011 at 5:52pm

Typically, we cease support for previous software versions when we release a new version. 

 

We plan to release a public beta of APM 2.0 tonight. The beta testing period typically lasts about six weeks, after when we release a final version. At that point, we will cease support of 1.02 and ask users to upgrade to the new version. 

 

This is the way most projects of this sort work, such as Arduino itself. Because the software is free and the hardware remains constant, there is no reason for all users not to upgrade to the latest version and allow the developers to focus on enhancing the latest code.

• ▶ Reply