Re: [Paparazzi-devel] New GPS for centimeter level precision GPS

[Karoly Molnar]

I am just thinking about the FPGA closed source code. If the operation capabilities of swift-nav would be a serious concern for the US government then the normal process is to classify the unit ITAR-controlled. Apparently, unless you are an army contractor you really want to avoid this because it restricts the market of the product significantly. So limiting the operation by FPGA code to avoid ITAR control might be a valid concern.
However I think a more realistic answer is that they just want to avoid mass-production of the hardware by other vendors. Actually this can be also justified, although I still consider the price too high. I am looking forward to seeing ublox and other GPS suppliers come out with their own DGPS or swiftnav-like solution in the coming year(s) for a more reasonable price.

Regards
Karoly

---

Date: Thu, 14 Nov 2013 09:47:40 +0100
From: address@hidden
To: address@hidden
Subject: Re: [Paparazzi-devel] New GPS for centimeter level precision GPS

Hello

The original email from swift nav:

Hi Marius,

We can provide you with the bill of materials if you like - however, at low quantities, getting the PCB and stencil made, sourcing the components, and assembling it yourself is probably going to be more expensive than buying it from us.

The FPGA boots from the M25P80 SPI Flash on board. We're intending to provide updates to the FPGA design as we develop it further, but for the most part the only thing that you should need to change while developing for the Piksi is the STM32 firmware. If you build the hardware yourself, you'll need to get a key for the FPGA design from us and program it into the SPI Flash - we'll provide that to you for free for small numbers of boards.

We're big fans of Paparazzi - we used to use it at Joby Energy while working on airborne wind turbines.

If you give me your email address I can email you a spreadsheet of the bill of materials.

Thanks,
Colin

Marius

I believe the main reason to have the FPGA source not available is that it otherwise is a security issue. They have to limit max speed and max height in order to be allowed to sell it. You can probably guess why...

On Nov 14, 2013 7:48 AM, "Marius Pfeffer" <address@hidden> wrote:

Hello

Found this:
http://lists.gnu.org/archive/html/paparazzi-devel/2013-10/msg00021.html

You need either a continous internet connection, a second telemetry link or the standard telemetry link to transmit the correction signals to the Piksi rover.
(look at the mailing list link)

Instead of xbees you can also use a raspberry pi with a umts stick and tunnel the serial port down to the gcs.
Should be easy, but mention most umts sticks have no own ip adress.

I talked to those guys from swift nav some months ago, all is open source except the fpga design file.
They were thinking about a open fpga architecture, i have no clue what happened with this idea.

Building some would be possible, but expensive !

6 Layer board design, stm32f4, max 2769, spartan 6 lx fpga ... very good hot air soldering oven required

If anyone is interested, i attached the BOM for the hardware revision v2.3.1
https://db.tt/L8sFwvMA

Regards
Marius

Am 14.11.2013 05:58 schrieb "Karoly Molnar" <address@hidden>:

This looks really interesting. However as I see it requires continuous internet data connection. Is there a option to use cell phone modems on UAVs both for this and for general bidirectional communication with the ground station? Or, to gateway the RTCM messages from the ground station to the plane as part of the uplink data?

Regards

Karoly

> Date: Wed, 13 Nov 2013 10:37:41 +0000
> From: address@hidden
> To: address@hidden
> Subject: Re: [Paparazzi-devel] New GPS for centimeter level precision GPS
>
> On 13 November 2013 09:55, Karoly Molnar <address@hidden> wrote:
> > Hi
> >
> > IMHO, USD 500 is rather steep price for a GPS.
> >
> > I have a second-hand information about a cheaper alternative. Once I had a
> > discussion with a vendor who is developing a data logger /live video capture
> > / gps capture solution for the automotive market. Due to the nature of this
> > business they are testing their solution with as much GPS recivers as they
> > find around the world. They have noted that the accuracy of the GPS
> > solutions vary significantly on each market. The most accurate GPS receivers
> > they have worked with were available for the South Korean internal market.
> > As they said, those were in the 10cm accuracy range. I did not explore this
> > further but if someone is interested, it might worth to check what is
> > available there and procure a few pieces for testing.
>
> My guess is they're using DGPS. There are a number of known base
> stations in South Korea.
>
> I looked into this briefly, and it seems like you can use DGPS with
> u-blox - all you have to do is send it RTCM messages. There's a list
> of transmitters here:
> http://www.ndblist.info/datamodes/worldDGPSdatabase.pdf (I'm guessing
> these send RTCM, but it doesn't say so explicitly).
>
> It also looks like you can get them over the 'net:
> http://software.rtcm-ntrip.org/. And possibly other places.
>
> This seems like something that wouldn't be too hard to add to the
> u-blox support.
>
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________ Paparazzi-devel mailing list address@hidden https://lists.nongnu.org/mailman/listinfo/paparazzi-devel