AIS not always accurate?

[phanakapan]

http://www.shipais.com/showship.php?mmsi=356172000
The vessel Salzgitter seems to be 'underway by sail'- given that it's a huge vehicle transport ship seems unlikely!

 

[matelot]

some feeds to ais will be direct (like position from a gps) and will be accurate. others will be by manual input and will be unreliable. you have to use common sense.

 

[alan17]

Rubish in rubish out

 

[Skyva_2]

From an NI 'Human Factors in AIS' study:

1193046 was used as an MMSI number by 26 vessels. Other instances of incorrect MMSI.

Ships with zero draught and/or zero length.

Reports of target swap and exchange of information suddenly between vessels.

3 similar high speed ferries variously described as 'Cargo', 'Passenger' and 'High Speed'.

Position 0 deg N 0 deg E

and much more...

 

[Danny Jo]

Some of you have seen this before, but I have received some bizarre headings via AIS. The figures that I'm interested in, though - position, Course Over Ground, and Speed Over Ground - are GPS derived and can be expected to be almost as reliable as GPS.

 

[peterb]

Knowing Urania, I suspect that the problem was simply a misplaced decimal point. Whether the point was lost at Urania's end or your's I wouldn't know, but the Royal Netherlands Navy are usually fairly careful about such things.

 

[nimbusgb]

What you have there is a bug in NASA's software. Nothing to do with AIS. Just NASA's implementation of AIS.

They are calculating bearing and distance using trigonometry, unless you get all your ducks in a row the algorithms will frequently trip up and display this sort of thing. Their routine should have a last minute sanity check before a bearing is displayed to make sure the range is within the range of 0 to 359 and adjust accordingly. 511-360 = 151

Easy mistake to make. Send NASA the photo and they may get around to modding the software.

 

[peterb]

[ QUOTE ]
They are calculating bearing and distance using trigonometry, unless you get all your ducks in a row the algorithms will frequently trip up and display this sort of thing. Their routine should have a last minute sanity check before a bearing is displayed to make sure the range is within the range of 0 to 359 and adjust accordingly. 511-360 = 151

[/ QUOTE ]

A heading of 151 with a COG of 52 and a SOG of 8.9 knots doesn't sound right, the boat would be going sideways. I prefer to bet on a heading of 51.1 with a missing decimal point.

 

[mikejames]

That's not an angle error, its NASA not decoding AIS properly. heading 511 is AIS code for vessel heading not available .

I bet they also give some ships going at 102.3 knots which is code for vessel speed not available.

There are a lot of little things like this where some codes are special cases which look silly if you simply decode the value.

So NASA need to refine their software.

 

[Danny Jo]

[ QUOTE ]
That's not an angle error, its NASA not decoding AIS properly. heading 511 is AIS code for vessel heading not available

[/ QUOTE ] That makes sense. Heading 511 has a kind of familiar feel to it, now that you mention it.

I tried, and failed to find a list of AIS codes. But I did come across this intriguing description of how a radio ham made his own AIS decoder http://rl.se/ais_eng

 

[BrianH]

Brian Lane's AIS parser site is invaluable in its message descriptors. See here.

Click on the 'Data Structures' tab at the top of the page.