Some GTFS-R Vehicle Positions Incorrect

Real-time Public Transport
1

I’m noticing that some of the coordinates returned by the GTFS-R vehicle position endpoint is a little strange.

I’ve been tracking a particular vehicle yesterday:

Vehicle ID: 4337.4338.4339.4340.4341.4342.4343.4344
Trip ID: 152T.435.123.60-20160801.A.8.42505972
Route ID: WL_1b

In sequential order, the GTFS-R feed returned the vehicle position to be:

Granville.Pendle Hill 3 Loc @ 1470063822
Granville.S18.1 Loc @ 1470063915
Granville.Toongabbie 3 Loc @ 1470063955

However, the coordinates returned for Granville.S18.1 Loc deviated significantly from the expected route:

Would this be because Granville.S18.1 Loc was incorrectly returned by the feed? or perhaps the coordinates for Granville.S18.1 Loc are placed on the wrong part of the track

Edit:

Pink is the route as per shapes.txt, blue is the route as plotted by the data returned by the GTFS-R feed. Star points are the individual nodes being plotted by the data returned by the GTFS-R feed.

2

Someone from our brains trust has advised that this is a known issue in the Sydney Trains’ system due to S18.1 being a train location name on two rail control systems.

3

Ah, I see. I suspect that’s a similar issue with Granville.S17.3 Loc?

4

could be…

there are 3 train locations that do it. it isn’t consistent and the impact is only brief - the train will jump off route then back on route 10-20 seconds later

1 Like
5

Thanks @yvonne.lee :slight_smile:

I think I’ve got a work around for the interim!

6

it’s also been raised with Sydney Trains to fix (apparently been fixed in the past… but something happened)

Thanks.