Composite Event Recognition for Maritime Monitoring Manolis - - PowerPoint PPT Presentation

▶
composite event recognition for maritime monitoring
SMART_READER_LITE
LIVE PREVIEW

Composite Event Recognition for Maritime Monitoring Manolis - - PowerPoint PPT Presentation

Jan 02, 2023 199 likes •448 views

Composite Event Recognition for Maritime Monitoring Manolis Pitsikalis 1 , Alexander Artikis 2 , 1 , Richard Dreo 3 ,Cyril Ray 3 , 4 , Elena Camossi 5 and Anne-Laure Jousselme 5 1 Institute of Informatics & Telecommunications, NCSR Demokritos,

slide-1
SLIDE 1

Composite Event Recognition for Maritime Monitoring

Manolis Pitsikalis1, Alexander Artikis2,1, Richard Dreo3,Cyril Ray3,4, Elena Camossi5 and Anne-Laure Jousselme5

1Institute of Informatics & Telecommunications, NCSR Demokritos, Athens, Greece 2Department of Maritime Studies, University of Piraeus, Greece 3Naval Academy Research Institute, Brest, France, 4Arts et Metiers ParisTech, France 5Centre for Maritime Research and Experimentation (CMRE), NATO, La Spezia, Italy

http://cer.iit.demokritos.gr

slide-2
SLIDE 2

Composite Event Recognition

INPUT ◮ RECOGNITION ◮ OUTPUT

Event Recognition System CE Definitions Streams of SDEs . . . . . . . . . . . . Recognised CEs . . . . . . . . . . . .

2

slide-3
SLIDE 3

Composite Event Recognition for Maritime Monitoring

3

slide-4
SLIDE 4

Composite Event Recognition Engine

Run-Time Event Calculus (RTEC):

  • Guides data-driven reasoning using domain-knowledge.
  • High-level language facilitating interaction with domain

experts.

  • Built-in rules for temporal reasoning.
  • Formal, declarative semantics.
  • Scalable to high-velocity data streams.
  • Direct routes to machine learning.

4

slide-5
SLIDE 5

Event Calculus

  • A logic programming language for representing and

reasoning about events and their effects.

  • Key components:

event (typically instantaneous). fluent: a property that may have different values at different points in time.

  • Built-in representation of inertia:

F = V holds at a particular time-point if F = V has been initiated by an event at some earlier time-point, and not terminated by another event in the meantime.

5

slide-6
SLIDE 6

Run-Time Event Calculus (RTEC)

Predicate Meaning happensAt(E, T) Event E occurs at time T initiatedAt(F = V , T) At time T a period of time for which F = V is initiated terminatedAt(F = V , T) At time T a period of time for which F = V is terminated holdsFor(F = V , I) I is the list of the maximal intervals for which F = V holds continuously holdsAt(F = V , T) The value of fluent F is V at time T union all([J1, . . . , Jn], I) I =(J1 ∪ . . . ∪ Jn) intersect all([J1, . . . , Jn], I) I =(J1 ∩ . . . ∩ Jn) relative complement all I = I ′ \ (J1 ∪ . . . ∪ Jn) (I ′, [J1, . . . , Jn], I) deadline[UE](F = V , T) F = V is terminated after T timepoints (Unless Extended)

6

slide-7
SLIDE 7

Run-Time Event Calculus (RTEC)

Predicate Meaning happensAt(E, T) Event E occurs at time T initiatedAt(F = V , T) At time T a period of time for which F = V is initiated terminatedAt(F = V , T) At time T a period of time for which F = V is terminated holdsFor(F = V , I) I is the list of the maximal intervals for which F = V holds continuously holdsAt(F = V , T) The value of fluent F is V at time T union all([J1, . . . , Jn], I) I =(J1 ∪ . . . ∪ Jn) intersect all([J1, . . . , Jn], I) I =(J1 ∩ . . . ∩ Jn) relative complement all I = I ′ \ (J1 ∪ . . . ∪ Jn) (I ′, [J1, . . . , Jn], I) deadline[UE](F = V , T) F = V is terminated after T timepoints (Unless Extended)

7

slide-8
SLIDE 8

Run-Time Event Calculus (RTEC)

Predicate Meaning happensAt(E, T) Event E occurs at time T initiatedAt(F = V , T) At time T a period of time for which F = V is initiated terminatedAt(F = V , T) At time T a period of time for which F = V is terminated holdsFor(F = V , I) I is the list of the maximal intervals for which F = V holds continuously holdsAt(F = V , T) The value of fluent F is V at time T union all([J1, . . . , Jn], I) I =(J1 ∪ . . . ∪ Jn) intersect all([J1, . . . , Jn], I) I =(J1 ∩ . . . ∩ Jn) relative complement all I = I ′ \ (J1 ∪ . . . ∪ Jn) (I ′, [J1, . . . , Jn], I) deadline[UE](F = V , T) F = V is terminated after T timepoints (Unless Extended)

8

slide-9
SLIDE 9

Run-Time Event Calculus (RTEC)

Predicate Meaning happensAt(E, T) Event E occurs at time T initiatedAt(F = V , T) At time T a period of time for which F = V is initiated terminatedAt(F = V , T) At time T a period of time for which F = V is terminated holdsFor(F = V , I) I is the list of the maximal intervals for which F = V holds continuously holdsAt(F = V , T) The value of fluent F is V at time T union all([J1, . . . , Jn], I) I =(J1 ∪ . . . ∪ Jn) intersect all([J1, . . . , Jn], I) I =(J1 ∩ . . . ∩ Jn) relative complement all I = I ′ \ (J1 ∪ . . . ∪ Jn) (I ′, [J1, . . . , Jn], I) deadline[UE](F = V , T) F = V is terminated after T timepoints (Unless Extended)

9

slide-10
SLIDE 10

CE Definitions in the RTEC

CE definition: initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIn1, T), happensAt(ET1, T), [conditions] [conditions] . . . . . . initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIni , T), happensAt(ETj , T), [conditions] [conditions] CE recognition:

time

10

slide-11
SLIDE 11

CE Definitions in the RTEC

CE definition: initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIn1, T), happensAt(ET1, T), [conditions] [conditions] . . . . . . initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIni , T), happensAt(ETj , T), [conditions] [conditions] CE recognition:

time

11

slide-12
SLIDE 12

CE Definitions in the RTEC

CE definition: initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIn1, T), happensAt(ET1, T), [conditions] [conditions] . . . . . . initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIni , T), happensAt(ETj , T), [conditions] [conditions] CE recognition:

time

12

slide-13
SLIDE 13

CE Definitions in the RTEC

CE definition: initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIn1, T), happensAt(ET1, T), [conditions] [conditions] . . . . . . initiatedAt(CE, T) ← terminatedAt(CE, T) ← happensAt(EIni , T), happensAt(ETj , T), [conditions] [conditions] CE recognition: holdsFor(CE, I)

time

13

slide-14
SLIDE 14

Maritime Patterns: Drifting

initiatedAt(drifting(Vessel), T) ← happensAt(velocity(Vessel, , CoG, TrHd), T), angleDiff (CoG, TrHd, Ad), threshold(vad, Vad), Ad > Vad, holdsAt(underWay(Vessel), T). terminatedAt(drifting(Vessel), T) ← happensAt(velocity(Vessel, , CoG, TrHd), T), angleDiff (CoG, TrHd, Ad), threshold(vad, Vad), Ad ≤ Vad. terminatedAt(drifting(Vessel), T) ← happensAt(end(underWay(Vessel)), T).

Critical Points True Heading Course over ground / AIS messages Trajectory

25 25 50 75 100 m

14

slide-15
SLIDE 15

Maritime Patterns: Trawling

initiatedAt(trawlingMovement(Vessel), T) ← happensAt(change in heading(Vessel), T), vesselType(Vessel, fishing), holdsAt(withinArea(Vessel, fishing), T). deadlineUE(trawlingMovement(Vessel), MinT). holdsFor(trawling(Vessel), I) ← holdsFor(trawlingMovement(Vessel), Itc), holdsFor(trawlSpeed(Vessel), It), intersect all([It, Itc], Ii), threshold(vtrawl, Vtrawl), intDurGreater(Ii, Vtrawl, I).

1 1 2 3 4 km

Critical points AIS messages Trajectory Fishing area

15

slide-16
SLIDE 16

Maritime Pattern Hierarchy

drifting highSpeedNC withinArea trawlSpeed anchoredOrMoored movingSpeed tuggingSpeed changingSpeed trawlingMovement loitering gap sarSpeed pilotBoarding stopped sarMovement underWay rendezVous sar trawling lowSpeed tugging

16

slide-17
SLIDE 17

Empirical Evaluation

17

slide-18
SLIDE 18

Empirical Evaluation

Attribute Brest, Europe France Period (months) 6 1 Vessels 5K 34K AIS signals 18M 55M Critical points 4.6M 17M Fishing areas 263 1K Natura 2000 areas 1K 6K Ports 222 2201 Spatio-temporal 811K 7M events

18

slide-19
SLIDE 19

Empirical Evaluation

Attribute Brest, Europe France Period (months) 6 1 Vessels 5K 34K AIS signals 18M 55M Critical points 4.6M 17M Fishing areas 263 1K Natura 2000 areas 1K 6K Ports 222 2201 Spatio-temporal 811K 7M events

19

slide-20
SLIDE 20

Precision based on expert feedback

Composite Event TP FP Precision anchoredOrMoored(Vessel) 3067 4 0.999 trawling(Vessel) 29 1.000 tugging(Vessel) 117 1.000 pilotBoarding(Vessel1, Vessel2) 80 1.000 rendezVous(Vessel1, Vessel2) 52 2 0.963 One month of the Brest dataset.

20

slide-21
SLIDE 21

Compression effects on accuracy

Composite Event Brest Europe F1-Score highSpeedNC(Vessel) 0.989 0.989 anchoredOrMoored(Vessel) 1.000 1.000 drifting(Vessel) 0.999

  • trawling(Vessel)

0.994 0.998 tugging(Vessel1, Vesssel2) 0.994 0.951 pilotBoarding(Vessel1, Vessel2) 1.000 1.000 rendezVous(Vessel1, Vessel2) 1.000 1.000 loitering(Vessel) 1.000 1.000 sar(Vessel) 0.998 0.988

21

slide-22
SLIDE 22

Performance Evaluation: Brest, France

2 4 8 16 50 100 150 200 Window size (hours) Average number of input entities (in thousands) Enriched AIS Stream Critical Point Stream 2 4 8 16 1 2 3 Window size (hours) Average recognition time (sec) Enriched AIS Stream Critical Point Stream 22

slide-23
SLIDE 23

Performance Evaluation: Europe

2 4 8 16 1 2 3 4 Window size (hours) Average number of input entities (in million) Enriched AIS Stream Critical Point Stream 2 4 8 16 5 10 15 Window size (hours) Average recognition time (min) Enriched AIS Stream Critical Point Stream 23

slide-24
SLIDE 24

Summary

Composite Event Recognition system for maritime monitoring:

  • with formal specifications of effective maritime patterns,
  • evaluated by domain experts using real data,
  • and proven to be capable of real-time CER.

Current work:

  • data fusion (AIS in conjuction with SAR images, radar etc),
  • and detection of dark targets.

The dataset of recognised composite events is available here: https://zenodo.org/record/2557290 Join us in the demo session or visit our site below: https://cer.iit.demokritos.gr/cermm

24