[acao-it] ACAO Tracking Protocol

[Daniele Orlandi]

On 06/07/2015 15:52, Thomas Sevaldrud wrote:
> Hi again,
> 
> 1. Interpolation:
> 
> * We need a fairly steady stream of fixes, with an interval of 5 seconds
> or less for the animations to look good.

Okay, most fixes will have a 1-second resolution.

> * We do our own interpolation on the client side, so we do not need the
> server to interpolate missing fixes. This will probably just do more
> harm than good.

Okay, seems fair.

> * We can handle missing fixes, but if these fixes are in a turn point
> sector, the scoring will break down.

The considerations I made in my previous email stand.

> Also if we lose fixes while the
> glider is circling or doing other maneuvers the interpolation will make
> the animated glider do strange things, since we have no way of knowing
> which way the glider is moving.

Let me do some measurement so that we will be able to understand how big
of a concern this would be.

> * Since the devices can not buffer and retransmit the missing fixes
> later, the main focus should be on getting the best possible coverage,
> especially near turn points to avoid outages.

Agreed.

> 2. Polling/Broadcasting
> 
> * We need a polling server which we can query for the full history of
> each flight in order to do competition scoring

Up and running already, just need some testing.

> * This polling server should associate the relevant flights for a
> competition day with the task/pilot info, preferrably through a SeeYou
> CUC file as specified in the existing Silent Wings/vPos protocol.

Okay, it's implemented already, I just have to implement a small GUI to
be able to upload the task CUC file when the task is set.

> * The server could also have a way of simply listing all flights that
> are active so we can test this without having to set up a competition
> for all the gliders.

Yes, this could be done easily. We already have a web-based
visualization interface, I could just add a way to see the tracker ID
that you could use to retrieve data for that flight.

> As soon as we get access to a running tracker server we will start
> testing, and hopefully we'll soon see how this works in practice.

Okay, I will give you the details as soon as possible.

Bye,

-- 
  Daniele Orlandi