Python Data Generation

When real or quality trajectory data is not available, it is possible to generate data sets given some constraints. Current methods for auto generation of trajectories are based on using airports or bounding boxes as the start and end points of the trajectory.

Generate Airport-based Trajectories

Tracktable includes data on many airports around the world. This data includes latitude/longitude, airport codes, and size rankings based on traffic. This data is available via the tracktable.info.airports package. The data generation package provides methods for generating direct trajectories between pairs of airports or between sets of airports. This can allow the user to simulate air traffic across the whole world or in a specific locality.

Important constraints for generating trajectories include speed, time between points, and the minimum number of points that make up the trajectory.

Generate Single Trajectory Between Airports

This example shows how to retrieve airport data using airport codes and creating a single trajectory.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
from datetime import datetime
from tracktable.examples.data.generators import generate_trajectories
from tracktable.info import airports

ABQ_AIRPORT = airports.airport_information("ABQ")
DEN_AIRPORT = airports.airport_information("DEN")
new_trajectory = generate_trajectories.generate_airport_trajectory(
                                          start_airport=ABQ_AIRPORT,
                                          end_airport=DEN_AIRPORT,
                                          start_time=datetime.now(),
                                          object_id='ABQ-DEN',
                                          desired_speed=400,
                                          seconds_between_points=60,
                                          minimum_num_points=10)

Generate Multiple Trajectories Between Lists Of Airports

This example shows how to retrieve a set of the ten largest airports and use it as input to the tracktable.examples.data.generators.generate_trajectories package. Alternatively, the tracktable.info.airports package could be used directly to get airports based on other criteria. The result will be a list of airports that start and end at random airports sampled from the given list.

1
2
3
4
5
6
7
8
9
from tracktable.examples.data.generators import generate_trajectories

ten_largest_airports = generate_trajectories.n_largest_airports(10)
new_trajectories = generate_trajectories.generate_random_airport_trajectories(
                                       start_airport_list=ten_largest_airports[:5],
                                       end_airport_list=ten_largest_airports[5:],
                                       num_paths=5,
                                       desired_speed=400,
                                       seconds_between_points=60)

Generate Multiple Trajectories Between Random Airports

This example shows the method call for generating trajectories between completely random airports. This is done by not defining lists for the starting and ending points.

1
2
3
4
5
6
7
8
from tracktable.examples.data.generators import generate_trajectories

new_trajectories = generate_trajectories.generate_random_airport_trajectories(
                                   start_airport_list=None,
                                   end_airport_list=[],
                                   num_paths=5,
                                   desired_speed=400,
                                   seconds_between_points=60)

Generate Trajectories Between Bounding Boxes

Bounding boxes can be created and used as the start and end points as part of trajectory generation. Points within the bounding boxes are randomly generated and used as the endpoints.

The constraints for creating these trajectories are identical to the ones used in the airport examples.

This example shows one method for creating the bounding boxes and using them as input to generate a list of 5 new trajectories.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
 from datetime import datetime
 from tracktable.examples.data.generators import generate_trajectories
 from tracktable.domain.terrestrial import TrajectoryPoint as TerrestrialTrajectoryPoint

 bbox_type = TerrestrialTrajectoryPoint.domain_classes['BoundingBox']
 starting_min_corner = TerrestrialTrajectoryPoint.domain_classes['BasePoint']()
 starting_max_corner = TerrestrialTrajectoryPoint.domain_classes['BasePoint']()
 ending_min_corner = TerrestrialTrajectoryPoint.domain_classes['BasePoint']()
 ending_max_corner = TerrestrialTrajectoryPoint.domain_classes['BasePoint']()

 albuquerque = TerrestrialTrajectoryPoint(-106.6504, 35.0844)
 san_francisco = TerrestrialTrajectoryPoint( -122.4194, 37.7749)
 atlanta = TerrestrialTrajectoryPoint(-84.42806, 33.636719)
 miami = TerrestrialTrajectoryPoint(-80.290556, 25.79325)

 starting_min_corner[0] = san_francisco[0]
 starting_min_corner[1] = albuquerque[1]
 starting_max_corner[0] = albuquerque[0]
 starting_max_corner[1] = san_francisco[1]

 ending_min_corner[0] = atlanta[0]
 ending_min_corner[1] = miami[1]
 ending_max_corner[0] = miami[0]
 ending_max_corner[1] = atlanta[1]

 starting_bbox = bbox_type(starting_min_corner, starting_max_corner)
 ending_bbox = bbox_type(ending_min_corner, ending_max_corner)

 new_trajectories = generate_trajectories.generate_bbox_trajectories(
                                                 starting_bbox,
                                                 ending_bbox,
                                                 5,
                                                 'BBOXTST',
                                                 start_time=datetime.now(),
                                                 desired_speed=400,
                                                 seconds_between_points=60,
                                                 minimum_num_points=10)