ST_Density function to calculate trajectory density - PolarDB

Calculates the density of a trajectory, defined as the number of trajectory points in the spatiotemporal neighborhood of each point.

Syntax

integer[] ST_Density(trajectory traj, float dr, interval dt);

Parameters

Parameter	Type	Description
`traj`	trajectory	The trajectory to evaluate.
`dr`	float	The spatial range of the neighborhood. Distance is calculated based on the spatial reference identifier (SRID) of the trajectory.
`dt`	interval	The time range of the neighborhood.

Description

ST_Density returns an integer array where each element is the density of the corresponding trajectory point — the count of sampling points (including the point itself) that fall within the spatiotemporal neighborhood.

A point P is in the neighborhood of point Q if both conditions hold:

The spatial distance between P and Q is less than dr. Distance is computed using the SRID of the trajectory.
The time difference between P and Q is less than dt.

Each point always counts itself, so the minimum density value is 1.

Example

SELECT ST_Density(
  st_makeTrajectory(
    'STPOINT'::leaftype,
    ARRAY[1::float8, 2, 3, 4, 5],
    ARRAY[2::float8, 10, 9, 8, 7],
    4326,
    ARRAY[
      '2010-01-01 11:30'::timestamp,
      '2010-01-01 11:31',
      '2010-01-01 11:32',
      '2010-01-01 11:33',
      '2010-01-01 11:34'
    ]
  ),
  400000,
  '10 minute'
);

Output:

 st_density
-------------
 {1,3,4,4,3}
(1 row)

The trajectory has 5 points with SRID 4326, timestamps from 11:30 to 11:34, a spatial neighborhood of 400,000 units, and a time neighborhood of 10 minutes. The result {1,3,4,4,3} gives the density for each point in order: point 1 has 1 neighbor (itself), points 2–4 are clustered together with higher densities, and point 5 has 3 neighbors.