Environmentally Safe Shipping

Efficient AIS Data Processing for

Environmentally Safe Shipping

Marios Vodas
1
, Nikos Pelekis
1
,
Yannis

Theodoridis
1
,

Cyril Ray
2
, Vangelis Karkaletsis
3
,
Sergios

Petridis
3
,

Anastasia Miliou
4


1
University of Piraeus

2
Naval Academy, France

3
NCSR “
Demokritos
”

4
Archipelago
–

Inst. of Marine Conservation

1

Outline

1.
Part I: Marine Transportation

2.
Part II: Automatic
Identification System (AIS
)

3.
Part
III
:
Objectives

4.
Part
IV: Methodology

5.
Part
V: Conclusion

2

I
.
MARITIME TRANSPORTATION

3

Safety (and Environmental) Issues


Ships, control centers and marine officers have to face many
security and safety problems due to:


Staff reduction, cognitive overload, human errors


Traffic increase (ports, maritime routes), dangerous contents


Terrorism, pirates


Technical faults (bad design, equipment breakdowns)


Bad weather


Etc.

4

MarineTraffic.com

HELCOME
AIS

IRENav

(NATO)

The Most Prominent Cause of Accidents


About 75
-
96% of marine casualties are caused, at least in
part, by some form of human error * :


88% of tanker accidents


79% of towing vessel groundings


96% of collisions


75% of fires and
explosions


Solution to such issues requires different levels of responses
taking into account :


People (activities)


Technology


Environment


Organisational factors


5

*
Rothblum

A.M.
(2006) “Human Error and Marine
Safety”, U.S. Coast Guard Research & Development
Center

Ways to Minimize Accidents


Level of education and practice
for

mariners


Work safety regulations (behaviour guidelines, normalised
onboard

equipments
)


Navigation and decision support systems providing real
-
time information,
predictions, alerts...


Integrate and use properly multiple and heterogeneous positioning
systems : AIS, ARPA, Long Range Identification System (LRIT),
Global
Maritime
Distress

and
Safety

System (GMDSS),
synthetic aperture radar,
airborne radar, satellite based sensors


Generalisation of vessel traffic monitoring, port control, search and rescue
systems, automatic
communications

6

Traffic Monitoring


7

Air
-
based

support

Human

and
semi
-
automatic
monitoring

On
-
demand

and on a
regular

basis

Remote Sensing support

Semi
-
automatic
monitoring

Every 2 to 6 hours

Sensor
-
based support

Almost automatic
analysis
and monitoring

Real
-
time

II
.
AUTOMATIC IDENTIFICATION
SYSTEM (AIS)

8

AIS Device


The Automatic Identification System
identifies and locates vessels at
distance


It includes an antenna, a transponder, a GPS receiver and additional
sensors (e.g., loch and gyrocompass)


It is a broadcast system based on VHF communications


It is able to operate in autonomous and continuous
mode


Ships fitted with AIS send navigation data to surrounding receivers (range
is about 50 km
)


Ships or maritime control centres on shore fitted with AIS receives
navigation data sent by surrounding
ships

9

→

AIS is mandatory (IMO) for big ships and
passengers’ boats

AIS Transmission Rate and Accuracy


AIS accuracy is defined as the largest distance the ship can
cover between two updates


The AIS broadcasts information with different rates of updates
depending on the ship’s current speed and manoeuvre


The IMO assumes that accuracy of embedded GPS is
10m

10

Vessel
behaviour

Time
between
updates

Accuracy (m)

Anchored

3 min

= 10 metres

Speed between 0
-
14 knots

12 s

Between 10 and 95 metres

Speed between 0
-
14 knots
and changing course

4 s

Between 10 and 40 metres

Speed between 14
-
23 knots

6 s

Between 55 and 80 metres

Speed between 14
-
23 knots
and changing course

2 s

Between 25 and 35 metres

Speed over 23 knots

3 s

> 45 metres

Speed over 23 knots and
changing course

2 s

> 35 metres

General update rules have been compared to reality: it appears that update rates are lower

AIS Data


The AIS provide location
-
based information on 2D routes, this
defining point
-
based 3D trajectories



Transmitted data include ship’s position and textual meta
-
information


Static
: ID number (MMSI), IMO code, ship name and type,
dimensions


Dynamic
: Position (Long,
Lat
), speed, heading, course over ground
(COG), rate of turn (ROT)


Route
-
based
: Destination, danger, estimated time of arrival (ETA)
and
draught

11

That is, an ordered
series of locations (X,Y,T)

of a given mobile object O
with T indicating the timestamp of the location (X,Y)

→
Time
does not exist in AIS frames : to be add by receivers

!AIVDM,1,1,,A,1Bwj:v0P1=1f75REQg>rPwv:0000,0*3B

III
. OBJECTIVES

12

Big AIS Data Processing for Environmentally
Safe Shipping


Objectives, based on
Archipelagos Institute of Marine
Conservation

requests, was to


Investigate factors which contribute most to the risk of a shipping
accident


Identify dangerous
areas


How : traffic database processing in order to address some
requirements /

queries

set by Archipelagos towards semi
-
quantitative risk analysis of shipping
traffic

13

→
Data
coming from
AIS


→
Application to the Aegean Sea

Typical Questions From Domain Experts


Calculate
average and minimum
distances

from shore or between
two
ships


Calculate the maximum
number of
ships

in the vicinity of another
ship


Find whether (and how many
times) a ship goes through
specified
areas

(e.g. narrow passages, biodiversity
boxes
)


Calculate the number of sharp
changes in ship
’
s
direction


Find
typical routes
vs.
outliers


etc. etc
.

14

Mediterranean Sea


European Maritime Safety Agency (EMSA) centralizes data
from EU states and provides them through a Web service



We worked on a dataset on
Mediterranean

sea

provided By
IMIS Hellas

(
a Greek IT company related to IMIS Global, collecting AIS data,
mariweb.gr
)


15

→
Data Volume is
100 million positions per month
, that is about 2300
positions per minutes

•
Focus on
Aegean sea :
3 days,
3
million
position records
(933
distinct ships
)


•
Full dataset is more than 2000 SQL
tables for a total of 2 TB covering 2,5
years of vessel activity

Two datasets are
available at Chorochronos.org interface (IMIS
3
days
and
AIS Brest)

Vessel Statistics

16

Country

Number of ships

Flag

of

Convenience

Greece

263

No

Panama

(
Republic

of)

112

Yes

Turkey

96

No

Malta

76

Yes

Liberia

(Republic

of)

32

Yes

Vincent

and

the

Grenadines

29

Yes

IV
. METHODOLOGY

17

Populating a Database


Relational database (
postgres

and
postgis
)


Data model based on AIS messages :
positions, ships and
trips


Parsing,
Integration
, error checking
filtering


Reconstructing trajectories from raw data
and feeding a trajectory
DB


Apply “simple” queries to answer experts
needs

18

“
What
is the (sub)trajectory of a ship during its
presence in an area
” ?

MOD Engine and Rule
-
Based Analysis


An integrated approach for maritime
situation awareness based on an inference
engine (drools)


The expert defines his rules according its needs
and objectives


The engine executes rules using the AIS
database

19


Hermes is a MOD engine providing extensible DBMS support
for trajectory data


Defines trajectory data type


SQL extensions at the logical level


Efficient indexing techniques at the physical level


Includes trajectory clustering support


Mixed top
-
down / bottom
-
up approach
involving an expert monitoring real
-
time
traffic on a touch table

http://
infolab.cs.unipi.gr/hermes

Methodology Steps


20

Cleaning
Filter
:


Wrong CRC


Duplicates
Decoding
AIS type
:


1
/
2
/
3


Position Report



5


Static and Voyage Related Data
Cleaning
Filter
:


Invalid MMSI


GPS Error



Hermes Loader


Degrees to Meters


Trajectory Update


Outputs Trajectories
Querying


Timeslice


Range


Temporal only


Spatial only


Spatio
-
Temporal


Nearest Neighbor
(
NN
)


wrt
.
a reference stati c object
(
point
/
segment
/
box
)


wrt
.
a reference trajectory
Advanced Querying


Pair
-
wise similarity queries


OD
-
Matrix


origin
/
destination are spatial vs
.
spatio
-
temporal boxes


Trajectory Clustering
Take the Maritime Environment Into
Account


The maritime domain is peculiar as there is no underlying
network but some maritime rules define predefined paths
and anchorage areas (polylines and polygons) that might
constrain a given trajectory



21

We added
official vector
chart
and
expert
-
defined areas
of
interest in the database


Coastlines


Starting
, ending,
passing, restricted
areas,
waiting zones


Regulations
and dangers (rocs, buoys,
seabed
)


…

S
-
57 ENC (Electronic Nautical
Chart)

Exploring the Data


Calculating trajectory aggregations and feeding a trajectory
data
warehouse


Performing OLAP analysis over aggregations (
eg
. O/D analysis
)


Running KDD techniques : frequent pattern analysis,
clustering, outlier detection, etc.


22

Cloud of locations

Association of points
coming from the same
source
-
destination set

Definition of a
route
and
qualifying of positions
at
each time

Qualifying of a new trajectory
compared to the identified route

Visualizing Trajectories and Patterns


23

←
space
-
time cube:
ship is late

space
-
time cube:
trajectory too far
on the right
→

s
peed
behaviour

frequent
patterns

→
Web
-
based

visualisation

using

Google Maps / Earth applications,
Openlayers

(OSM)

V. CONCLUSION

24

Some Open Questions

Q1. What kind of
storage

is appropriate for BIG volumes of
vessel traffic data?


Serial vs. parallel/distributed processing (e.g.
Hadoop
)


(batch vs. streaming) MOD engines?


What about indexing BIG mobility data
?

Q2. What kind of
analysis

on vessel traffic data makes
sense?


Analysis on current (location, speed, heading, …) vs. historical
information (trajectories)


Clusters (+ outliers), frequent patterns, next location prediction,
etc.


Exploit on previous knowledge to improve real
-
time
analysis

Q3. What kind of
visualization

is appropriate for
vessel traffic
data /

patterns


Current location vs. trajectory
-
based visual
analytics

25

Trajectory clustering

Frequent pattern

mining

Research Challenges on Data
–

Just a Few
Examples


Trajectory compression / simplification
: how to compress /
simplify trajectories keeping quality as high as possible
?


Semantic trajectory reconstruction
: how to extract semantics
from raw (GPS
-
based) trajectory data?


Trajectory sampling
: how to find a representative sample
among a trajectory dataset
?


Generating trajectories
by example: how to build large
synthetic datasets that simulate the
‘
behavior
’

of a small real
one
?


Etc
.

26

Questions

27