- iostream
- vector
- cmath
- limits
- numeric
- algorithm
- fstream
- string
- iostream
- ctime
- random
- matplotlib
- sys
- nbformat
The osm files can be generated from a osm.pbf, foundable at :
https://download.geofabrik.de/
Then, the tool osmosis is needed, foundable at :
https://github.com/openstreetmap/osmosis/releases/tag/0.48.3
Now, to extract the road network needed for Map Matching, execute :
osmosis --read-pbf belgium-latest.osm.pbf --tf accept-ways highway=* --used-node --bounding-box top=50.8562 left=4.3525 bottom=50.8126 right=4.3869 --write-xml file="brussels.osm"The bounding box can be changed.
To extract the rail network for the tramway, execute :
osmosis --read-pbf belgium-latest.osm.pbf --way-key-value keyValueList="railway.tram" --used-node --bounding-box top=50.8718 left=4.2967 bottom=50.7973 right=4.4096 --write-xml file="brussels_rail.osm"To extract the public transport stations, execute :
osmosis --read-pbf belgium-latest.osm.pbf --node-key-value keyValueList="public_transport.station,public_transport.stop_position" --bounding-box top=50.8718 left=4.2967 bottom=50.7973 right=4.4096 --write-xml file="bruxelles_stops.osm"The trajectory must be a GPX file, the road data are extracted from a OSM file
All the set up can be done in the main.cpp file :
- To change the OSM file : change path line 14
- To change the GPX file : change path line 16
- To change the number of points to match : change int line 22
- To change the output file name : chznge path line 33 The GPX and OSM files must not be modified from the standard format
g++ ./cpp/class.cpp ./cpp/context.cpp ./cpp/io.cpp ./cpp/main_context_city.cpp ./a.out python3 python/affichage.py output.txt The input file must be a csv file with the following structure
"n","id","t","x","y","label"
"1","219001125-3",2021-11-15 02:54:18,594767.946816762,6398600.51909148,"01-sailing"
The date is in format yyyy-mm-dd HH:MM:SS
The coordinates "x" and "y" are in meters (from any reference point).
The label can be "01-sailing", "02-fishing" or "03-sailing".
The id columns is ignored.
g++ ./cpp/class.cpp ./cpp/context.cpp ./cpp/io.cpp ./cpp/main_context_fishing.cpp ./a.out ./AIS_traj/traj/1.csv output.txt 1In the console is displayed the tuple
python3 python/affichagev1.py output.txt ./a.out ./AIS_traj/traj/1.csv output.txt 2In the console are displayed the 9 numbers that corresponds to the boxes in the table of the report. The 1st, 4th, 9th numbers must be the bigger and the other the lower.
python3 python/affichagev2.py output.txt All the set up can be done in the main_context_city.cpp file :
To compute or not the smoothing process, comment or not the line 38
To compute or not the light Map Matching post process, comment or not the line 46
A csv file, with the following structure
lon,lat,c1,c2,c3,date
4.36856078,50.85514968,0,0,0,129.59,2022.07.22 07:00:37+00,
The date is in format yyyy.mm.dd HH:MM:SS+00. The columns c1,c2,c3 are ignored.
If the file is not organised in the same way, the numbering of the columns can be changed from line 18 to 20.
The path can be changed line 21
The stations are in a osm file, see below how to generate it.
The path can be changed line 14
The railways are in a osm file, see below how to generate it.
The path can be changed line 13
g++ ./cpp/class.cpp ./cpp/context.cpp ./cpp/io.cpp ./cpp/main_context_city.cpp ./a.out The program write two txt files, by default output.txt and outputjson.txt. To display output.txt, execute :
python3 python/affichage.py output.txt outputjson.txt is a file that allows to see the result with geojson. To do so, copy-paste the file on the input console at :
https://www.convertcsv.com/csv-to-geojson.htm
Then, copy the output and paste at :
https://geojson.io/#map=2/20.0/0.0
The result will be similar as the figures in the report, with in addition the pink points representing the stations.