flyrobot wrote:Hi Ezio,
How to make change the map into satelite map at dashboard 3 ?
Thanks,
John
At the moment you can't. I haven't implemented it.
flyrobot wrote:Hi Ezio,
How to make change the map into satelite map at dashboard 3 ?
Thanks,
John
ezio wrote:Hi
Could you check this version https://dl.dropboxusercontent.com/u/805 ... iEZGUI.apk
If it works correctly I will publish it next week.
Bart
ezio wrote:Today when I was flying I noticed strange thing.
Distance to home in my app was always equals 0
I have I2C GPS and the MultiWii version was r1411.
I'm pretty sure that it worked before.
Also when I set the Home position, it was displayed on the map correctly (it is pulled from MultiWii) but quad returns to its original Home position.
Can anybody confirm this ?
Bart
zidlov wrote:ezio wrote:Today when I was flying I noticed strange thing.
Distance to home in my app was always equals 0
I have I2C GPS and the MultiWii version was r1411.
I'm pretty sure that it worked before.
Also when I set the Home position, it was displayed on the map correctly (it is pulled from MultiWii) but quad returns to its original Home position.
Can anybody confirm this ?
Bart
Hm, isn't it like that it's not possible to use the SetHome functions as well as the FollowMe with an I2C Gps? At least, that's why I ordered a MW AIO now, because I found out by accident that follow me didn't work and then read somewhere that it didn't work with an i2c gps. Would be great if I'm wrong, but I thought that's the way it is....
fryefryefrye wrote:Hello Ezio
For a long range communication, I'm using the following solution.
Android device+BT <==air==> BT module<-->uart<-->3DR modem <==air==> 3DR modem<-->uart<-->copter
Because the 3DR modem is not reliable communication channel.
Also the android device is just beside the BT module.
So, EZ-GUI will never know the copter is out of range. And the data on screen is stop updating.
Actually, when the copter is out of range, EZ-GUI is still sending request, but there is no response will be received.
Can you add some sound to indicate this situation is happened, or a time counter on the screen show how many second has been passed since last response from copter.
Thank you very much for your genius work.
fryefryefrye wrote:Hello Ezio
A small bug, when the heading in Dashboard 2 = 0, it will disappear.
hexjump wrote:fryefryefrye wrote:Hello Ezio
A small bug, when the heading in Dashboard 2 = 0, it will disappear.
Yes.
ezio wrote:fryefryefrye wrote:Hello Ezio
For a long range communication, I'm using the following solution.
Android device+BT <==air==> BT module<-->uart<-->3DR modem <==air==> 3DR modem<-->uart<-->copter
.
I will add it soon.
I had similar problems.
So just wait
hpcre wrote:Hello. Ezio,
I have a request. Currently, when editing PID values in-flight. It takes a few steps to do it. Like write the new value, click on menu, choose save to eeprom, then confirm choice in the confirmation box.
Can you please decrease these steps to just a click on a dedicated write button.
Thanks
fryefryefrye wrote:ezio wrote:fryefryefrye wrote:Hello Ezio
For a long range communication, I'm using the following solution.
Android device+BT <==air==> BT module<-->uart<-->3DR modem <==air==> 3DR modem<-->uart<-->copter
.
I will add it soon.
I had similar problems.
So just wait
Hello Ezio
What kind of long range communication system you are using? Is it running on full duplex or half-duplex?
I’m using a half-duplex industry wireless Uart running on 9600bps. Distance can be over 500m with small antenna.
I have tried running on 38400bps, only 100m can be reached.
Because the half-duplex and low baud rate, collision will happen. So I have to low down the refresh rate to 1Hz.
I’m thinking if you can have a “dynamic refresh rate”. It means send the next read command after receive last response. It will increase the refresh rate as fast as possible without cause any collision.
What do you think about it?
ezio wrote:hpcre wrote:Hello. Ezio,
I have a request. Currently, when editing PID values in-flight. It takes a few steps to do it. Like write the new value, click on menu, choose save to eeprom, then confirm choice in the confirmation box.
Can you please decrease these steps to just a click on a dedicated write button.
Thanks
Ok
I'll add a button below PID settings
ezio wrote:
Hi
I don't have long range system, I don't even have fpv stuffI was testing just for fun simple short range 433mhz trascivers.
"dynamic refresh rate" will be the best. But at the moment I have no idea how to implement it.
scrat wrote:I have MWii v2.2 on AIO Pro V1.1. And I'm using latest ezio app. When I click Other on the second screen I see there is an option for Magnetic Declination. Is this working with MWii 2.2 or I need to have some of the latest dev versions?
albertoLG wrote:Ezio, your software is great, thank you for your work, but I have a small problem or maybe a suggestion. When I try to set a new position hold on the map, seems difficult to click, map keep returning to the center of home position, and it's very difficult to click on it, need to try alot of times before I get the popup window, maybe I do something wrong? thanks again
Leon11t wrote:Hi Ezio. Your app uses integrated into Android voice synthesizer. He supports a limited number of languages. Ann example of Ukrainian language, synthesizer speaks only a numbers. It is very annoying! Could you able attach to the functions that are speaks a sound files. For example, when enabled GPS Hold, played file GPS_Hold.wav, Enabled Horizont mode played Horizont.wav or mp3 sound file. Anyone able to record their own personal sound files. Not necessari include these audio files in to the Play Market release. Additional files can be found in your GIT Hub repository. Is it possible implement this idea?
fryefryefrye wrote:ezio wrote:
Hi
I don't have long range system, I don't even have fpv stuffI was testing just for fun simple short range 433mhz trascivers.
"dynamic refresh rate" will be the best. But at the moment I have no idea how to implement it.
All half-duplex com system will have collision problem when you use the fixed time interval to send the request.
Especially when EZ-GUI wants to send position to copter. I have to click “set as position hold” button 2-3 times in the map, then the copter will start to move to the new hold position.
I have some idea about how to fix it. Just start to send the command just after received a reply from copter.
I’m a C++ developer, and had little experience in Android develop, and no experiences in open source software develop.
Maybe I can start my first open source software develop with your software on github.
Is that OK for you?
Alexinparis wrote:Hi,
I think waiting for ack for the next message is not a good solution.
It would just waste some free bandwidth waiting for messages.
A solution to avoid congestion could be a theoric bandwidth measurement based on MSP length
so that requests can't overflow the baud capacity.
+ a reasonable buffer to absorb spike.
ezio wrote:Well it is OK if your app will be worse then mine!
just kidding
ezio wrote:In my app it is not a truly fixed time interval send the request.
It works like this:
1 received data are processed
2 UI is updated, tts, etc...
3 standard requests are sent
4 app waits for the period set in config
6 go to 1
ezio wrote:BUT each time button is clicked (set PID, AUX, home position etc) the command is sent straight away and this is where the problem lies.
Alexinparis wrote:Hi,
I think waiting for ack for the next message is not a good solution.
It would just waste some free bandwidth waiting for messages.
A solution to avoid congestion could be a theoric bandwidth measurement based on MSP length
so that requests can't overflow the baud capacity.
+ a reasonable buffer to absorb spike.
fryefryefrye wrote:Talking about the free bandwidth wasted, I’m using a 433Mhz wireless Uart running on 9600bps. I must set refresh rate to 1000ms in EZ-GUI setting, otherwise all BOXNAME can not be get (may be caused by collision). Also some bandwidth was wasted in transfer needless data like Moto, Radio when in DashBoard2.
But long range is always half-duplex, unreliable and low-speed.
Alexinparis wrote:
I think 9600bps is too low for current GUI.
57600k is not too high to still have a good range and is enough to use fluently any GUI.
I think another confusion comes from the use of 3DR 433Mhz modules: the stock soft they use is not very good to handle a transparent UART
fryefryefrye wrote:ezio wrote:Well it is OK if your app will be worse then mine!
just kidding
I do not mean to start a new App, I was saying join you to update some mechanism in your APP. As you have fully understood my problem, I think it’s best to let you do it.
But I’m still struggling to make your source code working in my developer environment ezio wrote:In my app it is not a truly fixed time interval send the request.
It works like this:
1 received data are processed
2 UI is updated, tts, etc...
3 standard requests are sent
4 app waits for the period set in config
6 go to 1
I have understood that after review your code.
But there still a problem when the APP start and want to get all the BOXNAME.
In Unreliable com-systme, the first attempt to get BOXNAME may be failed. But you will not try it again. So the state in DashBoard 2 will not be show. Restart the App can solve it.ezio wrote:BUT each time button is clicked (set PID, AUX, home position etc) the command is sent straight away and this is where the problem lies.
That’s biggest problem I have. When I want to set new GPS-Hold Position, I must do it 2-3 times until it work. I’m gland you have known it. But the problem caused by Unreliable com-systme are still there. Check the GPS-Hold Position come back from copter is best .
ezio wrote:I'm thinking about the improvement of the communication for slow, half duplex systems.
And I have a questions:
Lets assume that we use half duplex communication
When GUI sends packet of requests, MultiWii starts to send a replies as soon as it gets first request. So MultiWii can start sending when others requests are still coming.
Am I right ?
How we can cope with this situation ?
Bart
Alexinparis wrote:
The current implementation is ask and answer but the answer is not needed to ask again, so you don't need to wait between several asks.
you can even group them is a single packet.
Code: Select all
requests = new int[] { MSP_STATUS, MSP_RAW_IMU, MSP_SERVO, MSP_MOTOR, MSP_RC, MSP_RAW_GPS, MSP_COMP_GPS, MSP_ALTITUDE, MSP_ATTITUDE, MSP_DEBUG };
sendRequestMSP(requestMSP(requests));
fryefryefrye wrote:ezio wrote:I'm thinking about the improvement of the communication for slow, half duplex systems.
And I have a questions:
Lets assume that we use half duplex communication
When GUI sends packet of requests, MultiWii starts to send a replies as soon as it gets first request. So MultiWii can start sending when others requests are still coming.
Am I right ?
How we can cope with this situation ?
Bart
As Alex have said:Alexinparis wrote:
The current implementation is ask and answer but the answer is not needed to ask again, so you don't need to wait between several asks.
you can even group them is a single packet.
And you have done like it:Code: Select all
requests = new int[] { MSP_STATUS, MSP_RAW_IMU, MSP_SERVO, MSP_MOTOR, MSP_RC, MSP_RAW_GPS, MSP_COMP_GPS, MSP_ALTITUDE, MSP_ATTITUDE, MSP_DEBUG };
sendRequestMSP(requestMSP(requests));
After a group of request sent out by GUI, MultiWii starts to send a group of replies. When MultiWii is finished and stop sending, the line is free and GUI can send the next group of request or a command like set Hold_postion.
You can not start sending while the line is busy when you are using half duplex systems.
If GUI can send the next group of request immediately after the line is free, we will get the maximal refresh rate at a given baud rate.
But we still need to think about if some data was lost in the air, what we should do.
Code: Select all
requests = new int[] { MSP_STATUS, MSP_RAW_IMU, MSP_SERVO, MSP_MOTOR, MSP_RC, MSP_RAW_GPS, MSP_COMP_GPS, MSP_ALTITUDE, MSP_ATTITUDE, MSP_DEBUG };
sendRequestMSP(requestMSP(requests));
linuxslate wrote:WalkOver;
Yes, the attitude indicator (artificial horizon) is backwards (vertically) in both Dashboard 2 and Dashboard 3.
I mentioned this to ezio in an email, and he wrote a bug report on github.
For casual readers Please Note:
We are not talking about the confusion some people have of left to right roll. This is a vertical (pitch) issue.
The attitude indicator in MultiWiiConf is correct. If pitching your aircraft (board) up/down, causes the correct reaction in MultiiWiiConf, you are OK to fly (in that respect).
scrat wrote:@ezio: Could you be so kind and post a link to the latest ez-gui (apk file) like is now on google play store. Thanks.
fryefryefrye wrote:But there still a problem when the APP start and want to get all the BOXNAME.
In Unreliable com-systme, the first attempt to get BOXNAME may be failed. But you will not try it again. So the state in DashBoard 2 will not be show. Restart the App can solve it.
ezio wrote:
Yes I know that
But when this is sent:Code: Select all
requests = new int[] { MSP_STATUS, MSP_RAW_IMU, MSP_SERVO, MSP_MOTOR, MSP_RC, MSP_RAW_GPS, MSP_COMP_GPS, MSP_ALTITUDE, MSP_ATTITUDE, MSP_DEBUG };
sendRequestMSP(requestMSP(requests));
multiwii doesn't wait for all requests be received. It starts to send reply as soon as it receives ie. MSP_STATUS request. So the line is busy because more requests are coming and at the same time multiwii starts to respond to already received requests.
ezio wrote:Hi
Today I added something very experimental. I will try to describe it here.
This feature allows to use any android map application to display a copter position. MultiWii EZ-GUI emulates Android's GPS.
copter's position should be displayed. Heading may not work in every app.
Let me know if it works.
Steve_In_Brussels wrote:I wondered if anybody else was getting this and what to do about it. I am running MW 2.1. EZ-GUI is truly great - well worth the Dashboard 3 too!!!
However, when I go into config and tick the "alt correction" box ("alt set to zero when arming"). First the alt stays at zero, lift off -- two metres, three metres happy happy. Then I start to get horrid wrong values, 4500 metres, minus 196 metres. When I look at dashboard 1, the alt is all over the place. Other parameters seem sensible.
Without all correct, I have nice steady altitudes, even if they are offset with respect to the ground.
Thoughts, comments or is this an odd bug (I have the latest version)? Nobody else seems to have the problem.