[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] Telemetry Over WiFi, through Panda Board

From: Gautier Hattenberger
Subject: Re: [Paparazzi-devel] Telemetry Over WiFi, through Panda Board
Date: Tue, 18 Jun 2013 17:53:39 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:10.0.12) Gecko/20130116 Icedove/10.0.12

Hi Cameron,

This FMS was more or less an interface to control the UAV from different sources (joystick, predefined signals like roll steps, onboard or ground based navigation computer). It is not really related to the GCS.
Also, Beth is a home made test bench that we have at Enac, with a Lisa/L, and not really used at the moment.


Le 18/06/2013 09:24, Cameron Lee a écrit :
So the Flight Management System is the Ground Segment including the GCS then? 

All your other answers make sense. Thanks Gautier.

As for our Wi-Fi setup Hector; this link is designed for in-flight transfer of still images from the plane to the ground. We use XTend radios for our autopilot telemetry. However, I've been working on support for multiple, redundant links ( and want to fly using this system. I'm hoping to have it working for demonstration or flight at the AUVSI SUAS competition at the end of this week. But I might not succeed. So the Wi-Fi telemetry doesn't need to be reliable. Having said that:

We use two Engenious Routers 802.11 g routers at 2.4 GHz. We've got a parabolic antenna at the ground station that someone points at the plane. We've tested with the airplane on the ground up to 3 Km with a throughput (bandwidth) of around 1 MB/s. At 6 Km, we were able to see the access point, but not connect to it reliably. In the lab, we got throughput up to around 2.5 MB/s.

During flights, we've had some loss of connection when the plane was within 500 m, but we think that this was due to the antenna pointer person not doing a good job (as in 90 degrees off). Nevertheless, we're far from considering this link reliable - we need lots more testing.

For RC we use a 2.4 GHz Spectrum DSMX transmitter. During the above range testing, we tested extensively for interference and didn't find any. At the 6 Km, the RC worked great even when the Wi-Fi was transmitting UDP packets at a full rate. The only interference we noticed was when the RC transmitter was positioned only a few meters in front of the parabolic antenna.

We've also carefully designed our antenna placement in the plane to try and make the antennas invisible to each other by making them co-linear or orthogonal.

The biggest RC interference issue we had was when we set up an access point at the ground station for other purposes. It was set to auto select a channel (and the plane's router was on a fixed frequency). But the new access point picked the exact same channel and blocked the signal. I suspect that it couldn't see the airplane's access point since it didn't have the parabolic antenna. We know to be very careful about this now.

I've got lots more details if you'd like - I wrote a whole report on this. I should probably make that publicly available somewhere. Let me know if you're interested and I'll put in the effort.


On Mon, Jun 17, 2013 at 2:13 AM, Gautier Hattenberger <address@hidden> wrote:
Hello Cameron,

Le 17/06/2013 09:07, Cameron Lee a écrit :

I'm looking into sending telemetry over WiFi using an embedded system on the aircraft such as the Panda Board, Beagle Board, a Gumstick Overo, or something similar. Specifically, I have a a Panda Board that listen's to the TWOG's UART output and want to send this data to the GCS over WiFi. Right now I only care about the one direction (downlink, aka telemetry, aka plane to ground). 

I'm looking at some code that I think does something similar in both software/airborne/fms and software/airborne/firmwares/beth. I've got a few questions though:

1. What does FMS stand for?
Flight Management System -> a really stupid name for this, we should change it if you want my opinion

2. What's the difference between overo_gcs_com.c and fms_gs_com.c? They seem like they do similar things. In particular, what's the difference between stuff in the fms folder and the stuff in the beth folder?
My colleague was working on this at the time of the Lisa/L. I forgot what the difference is. Probably not that much as you said.  You should search the makefiles in conf/firmwares, but you'll find which one are supposed to be compiled together.

3. I see GCS mentioned in many places in the code. For example: GCS_PORT. Is this for the telemetry, where data is sent to the ground segment including the GCS, but possibly also the messages agent and others? Or is this for datalink where data is sent from the GCS to control the plane? Or maybe both.
It was bi-directional telemetry/datalink over UDP if I recall correctly.

4. Does the existing code talk to the autopilot over UART? I don't see where to define which tty port this would be done on, so I'm wondering if maybe this isn't how it works.
No it was mainly using SPI link between MCU (stm32f1) and Overo board. Opening an UART should be easier...




Paparazzi-devel mailing list

Paparazzi-devel mailing list

Paparazzi-devel mailing list

reply via email to

[Prev in Thread] Current Thread [Next in Thread]