Introduction

AISDB was created to provide a complete set of tools to aid in the collection and processing of AIS data. AISDB can be used with either livestreaming AIS or historical raw AIS data files. The core data model behind AISDB is an SQLite database, and AISDB provides a Python interface to interact with the database; from database creation, querying, data processing, visualization, and exportation of data in CSV format. AISDB also provides tools for integrating AIS with environmental data in raster file formats. For example, AISDB provides convenience functions to download ocean bathymetric chart grids which may be used to query seafloor depth beneath each surface vessel position, although any raster data using longitude/latitude coordinate grid cells may be appended.

Index

Python Environment

Install with Pip

Install with Docker

Database Creation

From MERIDIAN’s data livestream

From historical AIS data files

Querying the database

Connect to the database

Get vessel trajectories from the database

Query a bounding box encapsulating a collection of zone polygons

Processing AIS messages

Voyage modelling

Data cleaning and MMSI deduplication

Interpolating, Geofencing, and filtering

Exporting as CSV

Integration with external metadata

Retrieve detailed vessel metadata from marinetraffic.com

Bathymetric charts

Rasters

Visualization

Display AIS data in the web interface

Data collection and sharing

Setting up an AIS receiving antenna

Sharing data to external networks

0. Python Environment

Requires Python version 3.8 or newer. Optionally requires SQLite (included in Python) or PostgresQL server (installed separately). The AISDB Python package can be installed using pip. It is recommended to install the package in a virtual Python environment such as venv.

python -m venv env_ais
source ./env_ais/*/activate
pip install aisdb

For information on installing AISDB from source code, see Installing from Source

AISDB may also be used with docker-compose to run as services, for more info see AISDB Docker docs. The Python code in the rest of this document can then be run in the new Python environment. When using an interpreter such as Jupyter, ensure that jupyter is installed in the same environment as AISDB.

1. Database Creation

Creating a database from live streaming data

A typical workflow for using AISDB requires a database of recorded AIS messages. The following code snippet demonstrates how to create a new database from MERIDIAN’s AIS data stream. with the argument stdout=True, the raw message input will be copied to stdout before it is decoded and added to the database. Also see the receiver api docs: aisdb.receiver.start_receiver().

from aisdb.receiver import start_receiver

start_receiver(connect_addr='aisdb.meridian.cs.dal.ca:9920', sqlite_dbpath='AIS.sqlitedb', stdout=True)

Creating a database from historical data files

An SQLite database file will be created at the specified database path dbpath. The source string will be stored in the database along with the decoded message data, making it easier to integrate data from multiple sources into the same database. A checksum of the first 1000 bytes from the input file will be stored to prevent processing the same data file twice. Checksum validation can be disabled by adding the argument skip_checksum=True. Decoding speed can be improved by placing the raw data files on a seperate hard drive from the database.

import aisdb

aisdb.decode_msgs(
  filepaths=['aisdb/tests/test_data_20210701.csv', 'aisdb/tests/test_data_20211101.nm4'],
  dbpath='AIS.sqlitedb',
  dbconn=aisdb.DBConn(),
  source='TESTING',
)

The decoder accepts raw AIS data in the .nm4 format, as long as a timestamp is included in the message header. For example:

\s:41925,c:1635731889,t:1635731965*66\!AIVDM,1,1,,,19NSRM@01v;inKaVqpGVUmN:00Rh,0*7C
\s:41925,c:1635731889,t:1635731965*66\!AIVDM,1,1,,,15Benl0000<P7Te`HQFVrU<804;`,0*39
\s:41925,c:1635731889,t:1635731965*66\!AIVDM,1,1,,,17`BO@7P@9;sbjwUDa7uSH:@00RQ,0*35

CSV formatted data files can also be used to create a database. When using CSV, the following header is expected:

MMSI,Message_ID,Repeat_indicator,Time,Millisecond,Region,Country,Base_station,Online_data,Group_code,Sequence_ID,Channel,Data_length,Vessel_Name,Call_sign,IMO,Ship_Type,Dimension_to_Bow,Dimension_to_stern,Dimension_to_port,Dimension_to_starboard,Draught,Destination,AIS_version,Navigational_status,ROT,SOG,Accuracy,Longitude,Latitude,COG,Heading,Regional,Maneuver,RAIM_flag,Communication_flag,Communication_state,UTC_year,UTC_month,UTC_day,UTC_hour,UTC_minute,UTC_second,Fixing_device,Transmission_control,ETA_month,ETA_day,ETA_hour,ETA_minute,Sequence,Destination_ID,Retransmit_flag,Country_code,Functional_ID,Data,Destination_ID_1,Sequence_1,Destination_ID_2,Sequence_2,Destination_ID_3,Sequence_3,Destination_ID_4,Sequence_4,Altitude,Altitude_sensor,Data_terminal,Mode,Safety_text,Non-standard_bits,Name_extension,Name_extension_padding,Message_ID_1_1,Offset_1_1,Message_ID_1_2,Offset_1_2,Message_ID_2_1,Offset_2_1,Destination_ID_A,Offset_A,Increment_A,Destination_ID_B,offsetB,incrementB,data_msg_type,station_ID,Z_count,num_data_words,health,unit_flag,display,DSC,band,msg22,offset1,num_slots1,timeout1,Increment_1,Offset_2,Number_slots_2,Timeout_2,Increment_2,Offset_3,Number_slots_3,Timeout_3,Increment_3,Offset_4,Number_slots_4,Timeout_4,Increment_4,ATON_type,ATON_name,off_position,ATON_status,Virtual_ATON,Channel_A,Channel_B,Tx_Rx_mode,Power,Message_indicator,Channel_A_bandwidth,Channel_B_bandwidth,Transzone_size,Longitude_1,Latitude_1,Longitude_2,Latitude_2,Station_Type,Report_Interval,Quiet_Time,Part_Number,Vendor_ID,Mother_ship_MMSI,Destination_indicator,Binary_flag,GNSS_status,spare,spare2,spare3,spare4

The aisdb.database.decoder.decode_msgs() function also accepts compressed .zip and .gz file formats as long as they can be decoded into either nm4 or CSV.

2. Querying the Database

Parameters for the database query can be defined using aisdb.database.dbqry.DBQuery. Iterate over rows returned from the database for each vessel with aisdb.database.dbqry.DBQuery.gen_qry(). Convert the results into generator yielding dictionaries with numpy arrays describing position vectors e.g. lon, lat, and time using aisdb.track_gen.TrackGen().

The following query will return vessel positions from the past 48 hours:

import aisdb
from datetime import datetime, timedelta

with aisdb.DBConn() as dbconn:
  qry = aisdb.DBQuery(
    dbconn=dbconn,
    dbpath='AIS.sqlitedb',
    callback=aisdb.database.sql_query_strings.in_timerange,
    start=datetime.utcnow() - timedelta(hours=48),
    end=datetime.utcnow(),
  )

  for vessel in aisdb.TrackGen(qry.gen_qry()):
      print(vessel)

A specific region can be queried for AIS data using aisdb.gis.Domain or one of its subclasses to define a collection of shapely polygon features. For this example, the domain contains a single bounding box polygon derived from a longitude/latitude coordinate pair and radial distance specified in meters. If multiple features are included in the domain object, the domain boundaries will encompass the convex hull of all features contained within.

with DBConn() as dbconn:
    domain = aisdb.DomainFromPoints(points=[(-63.6, 44.6),], radial_distances=[5000,])
    qry = aisdb.DBQuery(
        dbconn=dbconn,
        dbpath='AIS.sqlitedb',
        callback=aisdb.database.sqlfcn_callbacks.in_bbox_time_validmmsi,
        start=datetime.utcnow() - timedelta(hours=48),
        end=datetime.utcnow(),
        xmin=domain.boundary['xmin'],
        xmax=domain.boundary['xmax'],
        ymin=domain.boundary['ymin'],
        ymax=domain.boundary['ymax'],
    )

  for vessel in aisdb.TrackGen(qry.gen_qry()):
      print(vessel)

Additional query callbacks for filtering by region, timeframe, identifier, etc. can be found in aisdb.database.sql_query_strings and aisdb.database.sqlfcn_callbacks

3. Processing

Voyage Modelling

The generator described above can be input into a processing function, yielding modified results. For example, to model the activity of vessels on a per-voyage or per-transit basis, each voyage is defined as a continuous vector of vessel positions where the time between observed timestamps never exceeds a 24-hour period.

import aisdb
from datetime import datetime, timedelta

maxdelta = timedelta(hours=24)

with aisdb.DBConn() as dbconn:
  qry = aisdb.DBQuery(
    dbconn=dbconn,
    dbpath='AIS.sqlitedb',
    callback=aisdb.database.sql_query_strings.in_timerange,
    start=datetime.utcnow() - timedelta(hours=48),
    end=datetime.utcnow(),
  )

  tracks = aisdb.TrackGen(qry.gen_qry())
  track_segments = aisdb.split_timedelta(tracks, maxdelta)

  for segment in track_segments:
      print(segment)

Data cleaning and MMSI deduplication

A common issue with AIS is that the data is noisy, and databases may contain multiple vessels broadcasting with same identifier at the same time. The aisdb.denoising_encoder.encode_greatcircledistance() function uses an encoder to check the approximate distance between each vessel’s position, and then segments resulting vectors where a surface vessel couldn’t reasonably travel there using the most direct path, e.g. above 50 knots. A distance threshold and speed threshold are used as a hard limit on the maximum delta distance or delta time allowed between messages to be considered continuous. A score is computed for each position delta, with sequential messages in close proximity at shorter intervals given a higher score, calculated by haversine distance divided by elapsed time. Any deltas with a score not reaching the minimum threshold are considered as the start of a new segment. Finally, the beginning of each new segment is compared to the end of each existing segment with a matching vessel identifier, and if the delta exceeds the minimum score, the segments are concatenated. If multiple existing trajectories meet the minimum score threshold, the new segment will be concatenated the existing segment with the highest score.

Processing functions may be executed in sequence as a processing chain or pipeline, so after segmenting the individual voyages as shown above, results can be input into the encoder to effectively remove noise and correct for vessels with duplicate identifiers.

import aisdb
from datetime import datetime, timedelta

maxdelta = timedelta(hours=24)
distance_threshold = 200000  # meters
speed_threshold = 50  # knots
minscore = 1e-6

with aisdb.DBConn() as dbconn:
    qry = aisdb.DBQuery(
      dbconn=dbconn,
      dbpath='AIS.sqlitedb',
      callback=aisdb.database.sql_query_strings.in_timerange,
      start=datetime.utcnow() - timedelta(hours=48),
      end=datetime.utcnow(),
    )

    tracks = aisdb.TrackGen(qry.gen_qry())
    track_segments = aisdb.split_timedelta(tracks, maxdelta)
    tracks_encoded = aisdb.encode_greatcircledistance(track_segments, distance_threshold=distance_threshold, speed_threshold=speed_threshold, minscore=minscore)

In this second example, artificial noise is introduced into the tracks as a hyperbolic demonstration of the denoising capability. The resulting cleaned tracks are then displayed in the web interface.

import os
from datetime import datetime

import aisdb
from aisdb import DBQuery, DBConn
from aisdb.gis import DomainFromTxts

from dotenv import load_dotenv

load_dotenv()

dbpath = os.environ.get('EXAMPLE_NOISE_DB', 'AIS.sqlitedb')
trafficDBpath = os.environ.get('AISDBMARINETRAFFIC', 'marinetraffic.db')
domain = DomainFromTxts('EastCoast', folder=os.environ.get('AISDBZONES'))

start = datetime(2021, 7, 1)
end = datetime(2021, 7, 2)

default_boundary = {'xmin': -180, 'xmax': 180, 'ymin': -90, 'ymax': 90}


def random_noise(tracks, boundary=default_boundary):
    for track in tracks:
        i = 1
        while i < len(track['time']):
            track['lon'][i] *= track['mmsi']
            track['lon'][i] %= (boundary['xmax'] - boundary['xmin'])
            track['lon'][i] += boundary['xmin']
            track['lat'][i] *= track['mmsi']
            track['lat'][i] %= (boundary['ymax'] - boundary['ymin'])
            track['lat'][i] += boundary['ymin']
            i += 2
        yield track


with DBConn() as dbconn:
    vinfoDB = aisdb.webdata.marinetraffic.VesselInfo(trafficDBpath).trafficDB

    qry = DBQuery(
        dbconn=dbconn,
        dbpath=dbpath,
        start=start,
        end=end,
        callback=aisdb.database.sqlfcn_callbacks.in_bbox_time_validmmsi,
        **domain.boundary,
    )

    rowgen = qry.gen_qry(fcn=aisdb.database.sqlfcn.crawl_dynamic_static)

    tracks = aisdb.track_gen.TrackGen(rowgen, decimate=True)
    tracks = aisdb.webdata.marinetraffic.vessel_info(tracks, vinfoDB)
    tracks = random_noise(tracks, boundary=domain.boundary)
    tracks = aisdb.encode_greatcircledistance(tracks,
                                              distance_threshold=50000,
                                              minscore=1e-5,
                                              speed_threshold=50)

    if __name__ == '__main__':
        aisdb.web_interface.visualize(
            tracks,
            domain=domain,
            visualearth=True,
            open_browser=True,
        )

Interpolating, geofencing and filtering

Building on the above processing pipeline, the resulting cleaned trajectories can then be geofenced and filtered for results contained by atleast one domain polygon, and interpolated for uniformity.

# ...
    domain = aisdb.DomainFromPoints(points=[(-63.6, 44.6),], radial_distances=[5000,])
    tracks_filtered = aisdb.track_gen.fence_tracks(tracks_encoded, domain)
    tracks_interp = aisdb.interp_time(tracks_filtered, step=timedelta(minutes=15))

    for segment in track_segments:
        print(segment)

Additional processing functions can be found in the aisdb.track_gen module.

Exporting as CSV

The resulting processed voyage data can be exported in CSV format instead of being printed:

# ...
    aisdb.write_csv(tracks_interp, 'ais_24h_processed.csv')

4. Integration with external metadata

AISDB supports integration with external data sources such as bathymetric charts and other raster grids.

Bathymetric charts

To determine the approximate ocean depth at each vessel position, the aisdb.webdata.bathymetry module can be used.

import aisdb

# set the data storage directory
data_dir = './testdata/'

# download bathymetry grid from the internet
bathy = aisdb.webdata.bathymetry.Gebco(data_dir=data_dir)
bathy.fetch_bathymetry_grid()

Once the data has been downloaded, the Gebco() class may be used to append bathymetric data to tracks in the context of a TrackGen processing pipeline in the same manner as the processing functions described above.

# ...
     tracks = aisdb.TrackGen(qry.gen_qry())
     tracks_bathymetry = bathy.merge_tracks(tracks)

Also see aisdb.webdata.shore_dist.ShoreDist for determining approximate nearest distance to shore from vessel positions.

Rasters

Similarly, abritrary raster coordinate-gridded data may be appended to vessel tracks

# ...
     tracks = aisdb.TrackGen(qry.gen_qry())
     raster_path './GMT_intermediate_coast_distance_01d.tif'
     raster = aisdb.webdata.load_raster.RasterFile(raster_path)
     tracks = raster.merge_tracks(tracks, new_track_key="coast_distance")

Detailed metadata from marinetraffic.com

5. Visualization

AIS data from the database may be overlayed on a map such as the one shown above by using the aisdb.web_interface.visualize() function. This function accepts a generator of track dictionaries such as those output by aisdb.track_gen.TrackGen(). The color of each vessel track is determined by vessel type metadata.

import os
from datetime import datetime, timedelta

import aisdb
import aisdb.web_interface
from aisdb.tests.create_testing_data import (
    sample_database_file,
    random_polygons_domain,
)

domain = random_polygons_domain()

example_dir = 'testdata'
if not os.path.isdir(example_dir):
    os.mkdir(example_dir)

dbpath = os.path.join(example_dir, 'example_visualize.db')
months = sample_database_file(dbpath)
start = datetime(int(months[0][0:4]), int(months[0][4:6]), 1)
end = datetime(int(months[1][0:4]), int(months[1][4:6]) + 1, 1)


def color_tracks(tracks):
    ''' set the color of each vessel track using a color name or RGB value '''
    for track in tracks:
        track['color'] = 'red' or 'rgb(255,0,0)'
        yield track


with aisdb.SQLiteDBConn() as dbconn:
    qry = aisdb.DBQuery(
        dbconn=dbconn,
        dbpath=dbpath,
        start=start,
        end=end,
        callback=aisdb.sqlfcn_callbacks.valid_mmsi,
    )
    rowgen = qry.gen_qry()
    tracks = aisdb.track_gen.TrackGen(rowgen, decimate=False)
    tracks_segment = aisdb.track_gen.split_timedelta(tracks,
                                                     timedelta(weeks=4))
    tracks_colored = color_tracks(tracks_segment)

    if __name__ == '__main__':
        aisdb.web_interface.visualize(
            tracks_colored,
            domain=domain,
            visualearth=True,
            open_browser=True,
        )