Trajectory Data Warehouses - Data Warehouse Systems: Design and Implementation - page 479

SELECT ST Buffer(Route, 0.6, 0:10)
FROM Delivery
The result is composed of spatiotemporal cylindroids around the trajectories
of the trucks. As we have seen in Sect. 12.2 , we can use this operation
combined with the lifted intersection topological operation for testing
whether the routes of the two trucks intersect, as follows:
SELECT ST Intersects(ST Buffer(D1.Route, 0.6, 0:10),
ST Buffer(D2.Route, 0.6, 0:10))
FROM Delivery D1, Delivery D2
WHERE D1.TruckId = T1 AND D2.TruckId = T2
Finally, the following query computes the union of two moving points:
SELECT ST Union(D1.Route, D2.Route)
FROM Delivery D1, Delivery D2
WHERE D1.TruckId = T1 AND D2.TruckId = T2
The result of the query is given next:
T MULTIPOINT(
( Point(0 0) ' 08:00 ' , Point(2 2) ' 08:10 ' , Point(2 2) ' 08:25 ' )
( Point(1 0) ' 08:05 ' , Point(3 1) ' 08:15 ' , Point(3 2) ' 08:20 ' ))
For implementing temporal fields, we use temporal rasters defined by the
type T RASTER(P,Q) ,where P and Q are defined as above. We give next an
example that combines geometries and rasters. For this, we create a table
LandPlot describing parcels of land as follows:
CREATE TABLE LandPlot (
Id INT PRIMARY KEY,
Geom GEOMETRY(POLYGON),
SoilType RASTER,
Temp T RASTER(INSTANT,DATE) )
Here, the attribute Geom contains the geometry of the land plot, SoilType
contains a raster identifying the kinds of soil in the land plot, and Temp
contains a temporal raster reporting the temperatures in the land plot on a
daily basis. The query
SELECT L.Id, AtPeriod(L.Temp, PERIOD( ' 2012-04-01 ' , ' 2012-04-04 ' ))
FROM LandPlot L
returns for each land plot a temporal raster with the temperature for each
of the 3 days in the interval. Analogously, the following query returns for
each land plot the first day at which the temperature is reported and the
nontemporal raster of that day:
SELECT L.Id, InitialInstant(L.Temp), InitialValue(L.Temp)
FROM LandPlot L