Advertisements

Archive

PostGIS

Do you sometimes start writing an SQL query and around at line 50 you get the feeling that it might be getting out of hand? If so, it might be useful to start breaking it down into smaller chunks and wrap those up into custom functions. Never done that? Don’t despair! There’s an excellent PL/pgSQL tutorial on postgresqltutorial.com to get you started.

To get an idea of the basic structure of a PL/pgSQL function and to proof that PostGIS datatypes work just fine in this context, here’s a basic function that takes a trajectory geometry and outputs its duration, i.e. the difference between its last and first timestamp:

CREATE OR REPLACE FUNCTION AG_Duration(traj geometry) 
RETURNS numeric LANGUAGE 'plpgsql'
AS $BODY$ 
BEGIN
RETURN ST_M(ST_EndPoint(traj))-ST_M(ST_StartPoint(traj));
END; $BODY$;

My end goal for this exercise was to implement a function that takes a trajectory and outputs the stops along this trajectory. Commonly, a stop is defined as a long stay within an area with a small radius. This leads us to the following definition:

CREATE OR REPLACE FUNCTION AG_DetectStops(
   traj geometry, 
   max_size numeric, 
   min_duration numeric)
RETURNS TABLE(sequence integer, geom geometry) 
-- implementation follows here!

Note how this function uses RETURNS TABLE to enable it to return all the stops that it finds. To add a line to the output table, we need to assign values to the sequence and geom variables and then use RETURN NEXT.

Another reason to use PL/pgSQL is that it enables us to write loops. And loops I wanted for my stop detection function! Specifically, I wanted to go through all the points in the trajectory:

FOR pt IN SELECT (ST_DumpPoints(traj)).geom LOOP
-- here comes the magic!
END LOOP;

Eventually the function should go through the trajectory and identify all segments that stay within an area with max_size diameter for at least min_duration time. To test for the area size, we can use:

IF ST_MaxDistance(segment,pt) <= max_size THEN is_stop := true; 

Putting everything together, my current implementation looks like this:

CREATE OR REPLACE FUNCTION AG_DetectStops(
   traj geometry,
   max_size numeric,
   min_duration numeric)
RETURNS TABLE(sequence integer, geom geometry) 
LANGUAGE 'plpgsql'
AS $BODY$
DECLARE 
   pt geometry;
   segment geometry;
   is_stop boolean;
   previously_stopped boolean;
   stop_sequence integer;
   p1 geometry;
BEGIN
segment := NULL;
sequence := 0;
is_stop := false;
previously_stopped := false;
p1 := NULL;
FOR pt IN SELECT (ST_DumpPoints(traj)).geom LOOP
   IF segment IS NULL AND p1 IS NULL THEN 
      p1 := pt; 
   ELSIF segment IS NULL THEN 
      segment := ST_MakeLine(p1,pt); 
      p1 := NULL;
      IF ST_Length(segment) <= max_size THEN is_stop := true; END IF; 
   ELSE 
      segment := ST_AddPoint(segment,pt); 
      -- if we're in a stop, we want to grow the segment, otherwise we remove points to the specified min_duration 
      IF NOT is_stop THEN 
         WHILE ST_NPoints(segment) > 2 AND AG_Duration(ST_RemovePoint(segment,0)) >= min_duration LOOP
            segment := ST_RemovePoint(segment,0); 
         END LOOP;
      END IF;
      -- a stop is identified if the segment stays within a circle of diameter = max_size
      IF ST_Length(segment) <= max_size THEN is_stop := true; 
      ELSIF ST_Distance(ST_StartPoint(segment),pt) > max_size THEN is_stop := false;
      ELSIF ST_MaxDistance(segment,pt) <= max_size THEN is_stop := true; 
      ELSE is_stop := false; 
      END IF; 
      -- if we found the end of a stop, we need to check if it lasted long enough 
      IF NOT is_stop AND previously_stopped THEN 
         IF ST_M(ST_PointN(segment,ST_NPoints(segment)-1))-ST_M(ST_StartPoint(segment)) >= min_duration THEN
            geom := ST_RemovePoint(segment,ST_NPoints(segment)-1); 
            RETURN NEXT;
            sequence := sequence + 1;
            segment := NULL;
            p1 := pt;
         END IF;
      END IF;
   END IF;
   previously_stopped := is_stop;
END LOOP;
IF previously_stopped AND AG_Duration(segment) >= min_duration THEN 
   geom := segment; 
   RETURN NEXT; 
END IF;
END; $BODY$;

While this function is not really short, it’s so much more readable than my previous attempts of doing this in pure SQL. Some of the lines for determining is_stop are not strictly necessary but they do speed up processing.

Performance still isn’t quite where I’d like it to be. I suspect that all the adding and removing points from linestring geometries is not ideal. In general, it’s quicker to find shorter stops in smaller areas than longer stop in bigger areas.

Let’s test! 

Looking for a testing framework for PL/pgSQL, I found plpgunit on Github. While I did not end up using it, I did use its examples for inspiration to write a couple of tests, e.g.

CREATE OR REPLACE FUNCTION test.stop_at_beginning() RETURNS void LANGUAGE 'plpgsql'
AS $BODY$
DECLARE t0 integer; n0 integer;
BEGIN
WITH temp AS ( SELECT AG_DetectStops(
   ST_GeometryFromText('LinestringM(0 0 0, 0 0 1, 0.1 0.1 2, 2 2 3)'),
   1,1) stop 
)
SELECT ST_M(ST_StartPoint((stop).geom)), 
       ST_NPoints((stop).geom) FROM temp INTO t0, n0;	
IF t0 = 0 AND n0 = 3
   THEN RAISE INFO 'PASSED - Stop at the beginning of the trajectory';
   ELSE RAISE INFO 'FAILED - Stop at the beginning of the trajectory';
END IF;
END; $BODY$;

Basically, each test is yet another PL/pgSQL function that doesn’t return anything (i.e. returns void) but outputs messages about the status of the test. Here I made heavy use of the PERFORM statement which executes the provided function but discards the results:

Advertisements

In short: both writing trajectory queries as well as executing them is considerably faster using PostGIS trajectories (as LinestringM) rather than the commonly used point-based approach.

Here are a couple of examples to give you an impression of the differences.

Spoiler alert! Trajectory queries are up to 500 times faster than comparable point-based queries.

A quick look at indexing

In both cases, we have indexed the tracker id, geometry, and time columns to speed up query processing.

The trajectory table has 3 indexes

  • gist (time_range)
  • gist (track gist_geometry_ops_nd)
  • btree (tracker)

The point-based table has 4 indexes

  • gist (pt)
  • btree (trajectory_id)
  • btree (tracker)
  • btree (t)

Length

First, let’s see how to determine trajectory length for all observed moving objects (identified by a tracker id).

Using the point-based approach, we first need to ensure that the points are in the correct temporal order, create the lines, and finally sum up their length:

WITH ordered AS (
 SELECT trajectory_id, tracker, t, pt
 FROM geolife.trajectory_pt
 ORDER BY t
), tmp AS (
 SELECT trajectory_id, tracker, st_makeline(pt) traj
 FROM ordered 
 GROUP BY trajectory_id, tracker
)
SELECT tracker, round(sum(ST_Length(traj::geography)))
FROM tmp
GROUP BY tracker 
ORDER BY tracker

With trajectories, we can go right to computing lengths:

SELECT tracker, round(sum(ST_Length(track::geography)))
FROM geolife.trajectory_ext
GROUP BY tracker
ORDER BY tracker

On my test system, the trajectory query run time is 22.7 sec instead of 43.0 sec for the point-based approach:

Duration

Compared to trajectory length, duration is less complicated in the point-based approach:

WITH tmp AS (
 SELECT trajectory_id, tracker, min(t) start_time, max(t) end_time
 FROM geolife.trajectory_pt
 GROUP BY trajectory_id, tracker
)
SELECT tracker, sum(end_time - start_time)
FROM tmp
GROUP BY tracker
ORDER BY tracker

Still, the trajectory query is less complex and much faster at 31 ms instead of 6.0 sec:

SELECT tracker, sum(upper(time_range) - lower(time_range))
FROM geolife.trajectory_ext
GROUP BY tracker
ORDER BY tracker

Temporal filter

Extracting trajectories that occurred during a certain time frame is another common use case:

WITH tmp AS (
 SELECT trajectory_id, tracker, min(t) start_time, max(t) end_time
 FROM geolife.trajectory_pt
 GROUP BY trajectory_id, tracker
)
SELECT trajectory_id, tracker, start_time, end_time
FROM tmp
WHERE end_time > '2008-11-26 11:00'
AND start_time < '2008-11-26 15:00'
ORDER BY tracker

This point-based query takes 6.0 sec while the shorter trajectory query finishes in 12 ms:

SELECT id, tracker, time_range
FROM geolife.trajectory_ext
WHERE time_range && '[2008-11-26 11:00+1,2008-11-26 15:00+01]'::tstzrange

or equally fast (12 ms) by making use of the n-dimensional index:

WHERE track &&&	ST_Collect(
 ST_MakePointM(-180, -90, extract(epoch from '2008-11-26 11:00'::timestamptz)),
 ST_MakePointM(180, 90, extract(epoch from '2008-11-26 15:00'::timestamptz))
)

Spatial filter

Finally, of course, let’s have a look at spatial filters, for example, trajectories that start in a certain area:

WITH my AS ( 
 SELECT ST_Buffer(ST_SetSRID(ST_MakePoint(116.31894,39.97472),4326),0.0005) areaA
), tmp AS (
 SELECT trajectory_id, tracker, min(t) t 
 FROM geolife.trajectory_pt
 GROUP BY trajectory_id, tracker
)
SELECT distinct traj.tracker, traj.trajectory_id 
FROM tmp
JOIN geolife.trajectory_pt traj
ON tmp.trajectory_id = traj.trajectory_id AND traj.t = tmp.t
JOIN my
ON ST_Within(traj.pt, my.areaA)

This point-based query takes 6.0 sec while the shorter trajectory query finishes in 488 ms:

WITH my AS ( 
 SELECT ST_Buffer(ST_SetSRID(ST_MakePoint(116.31894, 39.97472),4326),0.0005) areaA
)
SELECT id, tracker, ST_AsText(track)
FROM geolife.trajectory_ext
JOIN my
ON areaA && track
AND ST_Within(ST_StartPoint(track), areaA)

For more generic “does this trajectory intersect another geometry”, the points can also be aggregated to a linestring on the fly but that takes 21.9 sec:

I’ll be presenting more work on PostGIS trajectories at GI_Forum in Salzburg in July. In the talk, I’ll also have a look at the custom PG-Trajectory datatype. Here’s the full open-access paper:

Graser, A. (2018) Evaluating Spatio-temporal Data Models for Trajectories in PostGIS Databases. GI_Forum ‒ Journal of Geographic Information Science, 1-2018, 16-33. DOI: 10.1553/giscience2018_01_s16.


This post is part of a series. Read more about movement data in GIS.

MarineCadastre.gov is a great source for AIS data along the US coast. Their data formats and tools though are less open. Luckily, GDAL – and therefore QGIS – can read ESRI File Geodatabases (.gdb).

MarineCadastre.gov also offer a Track Builder script that creates lines out of the broadcast points. (It can also join additional information from the vessel and voyage layers.) We could reproduce the line creation step using tools such as Processing’s Point to path but this post will show how to create PostGIS trajectories instead.

First, we have to import the points into PostGIS using either DB Manager or Processing’s Import into PostGIS tool:

Then we can create the trajectories. I’ve opted to create a materialized view:

The first part of the query creates a temporary table called ptm (short for PointM). This step adds time stamp information to each point. The second part of the query then aggregates these PointMs into trajectories of type LineStringM.

CREATE MATERIALIZED VIEW ais.trajectory AS
 WITH ptm AS (
   SELECT b.mmsi,
     st_makepointm(
       st_x(b.geom), 
       st_y(b.geom), 
       date_part('epoch', b.basedatetime)
     ) AS pt,
     b.basedatetime t
   FROM ais.broadcast b
   ORDER BY mmsi, basedatetime
 )
 SELECT row_number() OVER () AS id,
   st_makeline(ptm.pt) AS st_makeline,
   ptm.mmsi,
   min(ptm.t) AS min_t,
   max(ptm.t) AS max_t
 FROM ptm
 GROUP BY ptm.mmsi
WITH DATA;

The trajectory start and end times (min_t and max_t) are optional but they can help speed up future queries.

One of the advantages of creating trajectory lines is that they render many times faster than the original points.

Of course, we end up with some artifacts at the border of the dataset extent. (Files are split by UTM zone.) Trajectories connect the last known position before the vessel left the observed area with the position of reentry. This results, for example, in vertical lines which you can see in the bottom left corner of the above screenshot.

With the trajectories ready, we can go ahead and start exploring the dataset. For example, we can visualize trajectory speed and/or create animations:

Purple trajectory segments are slow while green segments are faster

We can also perform trajectory analysis, such as trajectory generalization:

This is a first proof of concept. It would be great to have a script that automatically fetches the datasets for a specified time frame and list of UTM zones and loads them into PostGIS for further processing. In addition, it would be great to also make use of the information in the vessel and voyage tables, thus splitting up trajectories into individual voyages.


This post is part of a series. Read more about movement data in GIS.

This is a guest post by Chris Kohler .

Introduction:

This guide provides step-by-step instructions to produce drive-time isochrones using a single vector shapefile. The method described here involves building a routing network using a single vector shapefile of your roads data within a Virtual Box. Furthermore, the network is built by creating start and end nodes (source and target nodes) on each road segment. We will use Postgresql, with PostGIS and Pgrouting extensions, as our database. Please consider this type of routing to be fair, regarding accuracy, as the routing algorithms are based off the nodes locations and not specific addresses. I am currently working on an improved workflow to have site address points serve as nodes to optimize results. One of the many benefits of this workflow is no financial cost to produce (outside collecting your roads data). I will provide instructions for creating, and using your virtual machine within this guide.

Steps:–Getting Virtual Box(begin)–

Intro 1. Download/Install Oracle VM(https://www.virtualbox.org/wiki/Downloads)

Intro 2. Start the download/install OSGeo-Live 11(https://live.osgeo.org/en/overview/overview.html).

Pictures used in this workflow will show 10.5, though version 11 can be applied similarly. Make sure you download the version: osgeo-live-11-amd64.iso. If you have trouble finding it, here is the direct link to the download (https://sourceforge.net/projects/osgeo-live/files/10.5/osgeo-live-10.5-amd64.iso/download)
Intro 3. Ready for virtual machine creation: We will utilize the downloaded OSGeo-Live 11 suite with a virtual machine we create to begin our workflow. The steps to create your virtual machine are listed below. Also, here are steps from an earlier workshop with additional details with setting up your virtual machine with osgeo live(http://workshop.pgrouting.org/2.2.10/en/chapters/installation.html).

1.  Create Virutal Machine: In this step we begin creating the virtual machine housing our database.

Open Oracle VM VirtualBox Manager and select “New” located at the top left of the window.

VBstep1

Then fill out name, operating system, memory, etc. to create your first VM.

vbstep1.2

2. Add IDE Controller:  The purpose of this step is to create a placeholder for the osgeo 11 suite to be implemented. In the virtual box main window, right-click your newly-created vm and open the settings.

vbstep2

In the settings window, on the left side select the storage tab.

Find “adds new storage controller button located at the bottom of the tab. Be careful of other buttons labeled “adds new storage attachment”! Select “adds new storage controller button and a drop-down menu will appear. From the top of the drop-down select “Add IDE Controller”.

vbstep2.2

vbstep2.3

You will see a new item appear in the center of the window under the “Storage Tree”.

3.  Add Optical Drive: The osgeo 11 suite will be implemented into the virtual machine via an optical drive. Highlight the new controller IDE you created and select “add optical drive”.

vbstep3

A new window will pop-up and select “Choose Disk”.

vbstep3.2

Locate your downloaded file “osgeo-live 11 amd64.iso” and click open. A new object should appear in the middle window under your new controller displaying “osgeo-live-11.0-amd64.iso”.

vbstep3.3

Finally your virtual machine is ready for use.
Start your new Virtual Box, then wait and follow the onscreen prompts to begin using your virtual machine.

vbstep3.4

–Getting Virtual Box(end)—

4. Creating the routing database, and both extensions (postgis, pgrouting): The database we create and both extensions we add will provide the functions capable of producing isochrones.

To begin, start by opening the command line tool (hold control+left-alt+T) then log in to postgresql by typing “psql -U user;” into the command line and then press Enter. For the purpose of clear instruction I will refer to database name in this guide as “routing”, feel free to choose your own database name. Please input the command, seen in the figure below, to create the database:

CREATE DATABASE routing;

You can use “\c routing” to connect to the database after creation.

step4

The next step after creating and connecting to your new database is to create both extensions. I find it easier to take two-birds-with-one-stone typing “psql -U user routing;” this will simultaneously log you into postgresql and your routing database.

When your logged into your database, apply the commands below to add both extensions

CREATE EXTENSION postgis;
CREATE EXTENSION pgrouting;

step4.2

step4.3

5. Load shapefile to database: In this next step, the shapefile of your roads data must be placed into your virtual machine and further into your database.

My method is using email to send myself the roads shapefile then download and copy it from within my virtual machines web browser. From the desktop of your Virtual Machine, open the folder named “Databases” and select the application “shape2pgsql”.

step5

Follow the UI of shp2pgsql to connect to your routing database you created in Step 4.

step5.2

Next, select “Add File” and find your roads shapefile (in this guide we will call our shapefile “roads_table”) you want to use for your isochrones and click Open.

step5.3

Finally, click “Import” to place your shapefile into your routing database.

6. Add source & target columns: The purpose of this step is to create columns which will serve as placeholders for our nodes data we create later.

There are multiple ways to add these columns into the roads_table. The most important part of this step is which table you choose to edit, the names of the columns you create, and the format of the columns. Take time to ensure the source & target columns are integer format. Below are the commands used in your command line for these functions.

ALTER TABLE roads_table ADD COLUMN "source" integer;
ALTER TABLE roads_table ADD COLUMN "target" integer;

step6

step6.2

7. Create topology: Next, we will use a function to attach a node to each end of every road segment in the roads_table. The function in this step will create these nodes. These newly-created nodes will be stored in the source and target columns we created earlier in step 6.

As well as creating nodes, this function will also create a new table which will contain all these nodes. The suffix “_vertices_pgr” is added to the name of your shapefile to create this new table. For example, using our guide’s shapefile name , “roads_table”, the nodes table will be named accordingly: roads_table_vertices_pgr. However, we will not use the new table created from this function (roads_table_vertices_pgr). Below is the function, and a second simplified version, to be used in the command line for populating our source and target columns, in other words creating our network topology. Note the input format, the “geom” column in my case was called “the_geom” within my shapefile:

pgr_createTopology('roads_table', 0.001, 'geom', 'id',
 'source', 'target', rows_where := 'true', clean := f)

step7

Here is a direct link for more information on this function: http://docs.pgrouting.org/2.3/en/src/topology/doc/pgr_createTopology.html#pgr-create-topology

Below is an example(simplified) function for my roads shapefile:

SELECT pgr_createTopology('roads_table', 0.001, 'the_geom', 'id')

8. Create a second nodes table: A second nodes table will be created for later use. This second node table will contain the node data generated from pgr_createtopology function and be named “node”. Below is the command function for this process. Fill in your appropriate source and target fields following the manner seen in the command below, as well as your shapefile name.

To begin, find the folder on the Virtual Machines desktop named “Databases” and open the program “pgAdmin lll” located within.

step8

Connect to your routing database in pgAdmin window. Then highlight your routing database, and find “SQL” tool at the top of the pgAdmin window. The tool resembles a small magnifying glass.

step8.2

We input the below function into the SQL window of pgAdmin. Feel free to refer to this link for further information: (https://anitagraser.com/2011/02/07/a-beginners-guide-to-pgrouting/)

CREATE TABLE node AS
   SELECT row_number() OVER (ORDER BY foo.p)::integer AS id,
          foo.p AS the_geom
   FROM (     
      SELECT DISTINCT roads_table.source AS p FROM roads_table
      UNION
      SELECT DISTINCT roads_table.target AS p FROM roads_table
   ) foo
   GROUP BY foo.p;

step8.3

  1.  Create a routable network: After creating the second node table from step 8,  we will combine this node table(node) with our shapefile(roads_table) into one, new, table(network) that will be used as the routing network. This table will be called “network” and will be capable of processing routing queries.  Please input this command and execute in SQL pgAdmin tool as we did in step 8. Here is a reference for more information:(https://anitagraser.com/2011/02/07/a-beginners-guide-to-pgrouting/)   

step8.2

 

CREATE TABLE network AS
   SELECT a.*, b.id as start_id, c.id as end_id
   FROM roads_table AS a
      JOIN node AS b ON a.source = b.the_geom
      JOIN node AS c ON a.target = c.the_geom;

step9.2

10. Create a “noded” view of the network:  This new view will later be used to calculate the visual isochrones in later steps. Input this command and execute in SQL pgAdmin tool.

CREATE OR REPLACE VIEW network_nodes AS 
SELECT foo.id,
 st_centroid(st_collect(foo.pt)) AS geom 
FROM ( 
  SELECT network.source AS id,
         st_geometryn (st_multi(network.geom),1) AS pt 
  FROM network
  UNION 
  SELECT network.target AS id, 
         st_boundary(st_multi(network.geom)) AS pt 
  FROM network) foo 
GROUP BY foo.id;

step10

11.​ Add column for speed:​ This step may, or may not, apply if your original shapefile contained a field of values for road speeds.

In reality a network of roads will typically contain multiple speed limits. The shapefile you choose may have a speed field, otherwise the discrimination for the following steps will not allow varying speeds to be applied to your routing network respectfully.

If values of speed exists in your shapefile we will implement these values into a new field, “traveltime“, that will show rate of travel for every road segment in our network based off their geometry. Firstly, we will need to create a column to store individual traveling speeds. The name of our column will be “traveltime” using the format: ​double precision.​ Input this command and execute in the command line tool as seen below.

ALTER TABLE network ADD COLUMN traveltime double precision;

step11

Next, we will populate the new column “traveltime” by calculating traveling speeds using an equation. This equation will take each road segments geometry(shape_leng) and divide by the rate of travel(either mph or kph). The sample command I’m using below utilizes mph as the rate while our geometry(shape_leng) units for my roads_table is in feet​. If you are using either mph or kph, input this command and execute in SQL pgAdmin tool. Below further details explain the variable “X”.

UPDATE network SET traveltime = shape_leng / X*60

step11.2

How to find X​, ​here is an example​: Using example 30 mph as rate. To find X, we convert 30 miles to feet, we know 5280 ft = 1 mile, so we multiply 30 by 5280 and this gives us 158400 ft. Our rate has been converted from 30 miles per hour to 158400 feet per hour. For a rate of 30 mph, our equation for the field “traveltime”  equates to “shape_leng / 158400*60″. To discriminate this calculations output, we will insert additional details such as “where speed = 30;”. What this additional detail does is apply our calculated output to features with a “30” value in our “speed” field. Note: your “speed” field may be named differently.

UPDATE network SET traveltime = shape_leng / 158400*60 where speed = 30;

Repeat this step for each speed value in your shapefile examples:

UPDATE network SET traveltime = shape_leng / X*60 where speed = 45;
UPDATE network SET traveltime = shape_leng / X*60 where speed = 55;

The back end is done. Great Job!

Our next step will be visualizing our data in QGIS. Open and connect QGIS to your routing database by right-clicking “PostGIS” in the Browser Panel within QGIS main window. Confirm the checkbox “Also list tables with no geometry” is checked to allow you to see the interior of your database more clearly. Fill out the name or your routing database and click “OK”.

If done correctly, from QGIS you will have access to tables and views created in your routing database. Feel free to visualize your network by drag-and-drop the network table into your QGIS Layers Panel. From here you can use the identify tool to select each road segment, and see the source and target nodes contained within that road segment. The node you choose will be used in the next step to create the views of drive-time.

12.Create views​: In this step, we create views from a function designed to determine the travel time cost. Transforming these views with tools will visualize the travel time costs as isochrones.

The command below will be how you start querying your database to create drive-time isochrones. Begin in QGIS by draging your network table into the contents. The visual will show your network as vector(lines). Simply select the road segment closest to your point of interest you would like to build your isochrone around. Then identify the road segment using the identify tool and locate the source and target fields.

step12

step12.2

Place the source or target field value in the below command where you see ​VALUE​, in all caps​.

This will serve you now as an isochrone catchment function for this workflow. Please feel free to use this command repeatedly for creating new isochrones by substituting the source value. Please input this command and execute in SQL pgAdmin tool.

*AT THE BOTTOM OF THIS WORKFLOW I PROVIDED AN EXAMPLE USING SOURCE VALUE “2022”

CREATE OR REPLACE VIEW "​view_name" AS 
SELECT di.seq, 
       di.id1, 
       di.id2, 
       di.cost, 
       pt.id, 
       pt.geom 
FROM pgr_drivingdistance('SELECT
     gid::integer AS id, 
     Source::integer AS source, 
     Target::integer AS target,                                    
     Traveltime::double precision AS cost 
       FROM network'::text, ​VALUE::bigint, 
    100000::double precision, false, false)
    di(seq, id1, id2, cost)
JOIN network_nodes pt ON di.id1 = pt.id;

step12.3

13.Visualize Isochrone: Applying tools to the view will allow us to adjust the visual aspect to a more suitable isochrone overlay.

​After creating your view, a new item in your routing database is created, using the “view_name” you chose. Drag-and-drop this item into your QGIS LayersPanel. You will see lots of small dots which represent the nodes.

In the figure below, I named my view “take1“.

step13

Each node you see contains a drive-time value, “cost”, which represents the time used to travel from the node you input in step 12’s function.

step13.2

Start by installing the QGIS plug-in Interpolation” by opening the Plugin Manager in QGIS interface.

step13.3

Next, at the top of QGIS window select “Raster” and a drop-down will appear, select “Interpolation”.

step13.4

 

A new window pops up and asks you for input.

step13.5

Select your “​view”​ as the​ vector layer​, select ​”cost​” as your ​interpolation attribute​, and then click “Add”.

step13.6

A new vector layer will show up in the bottom of the window, take care the type is Points. For output, on the other half of the window, keep the interpolation method as “TIN”, edit the ​output file​ location and name. Check the box “​Add result to project​”.

Note: decreasing the cellsize of X and Y will increase the resolution but at the cost of performance.

Click “OK” on the bottom right of the window.

step13.7

A black and white raster will appear in QGIS, also in the Layers Panel a new item was created.

step13.8

Take some time to visualize the raster by coloring and adjusting values in symbology until you are comfortable with the look.

step13.9

step13.10

14. ​Create contours of our isochrone:​ Contours can be calculated from the isochrone as well.

Find near the top of QGIS window, open the “Raster” menu drop-down and select Extraction → Contour.

step14

Fill out the appropriate interval between contour lines but leave the check box “Attribute name” unchecked. Click “OK”.

step14.2

step14.3

15.​ Zip and Share:​ Find where you saved your TIN and contours, compress them in a zip folder by highlighting them both and right-click to select “compress”. Email the compressed folder to yourself to export out of your virtual machine.

Example Isochrone catchment for this workflow:

CREATE OR REPLACE VIEW "2022" AS 
SELECT di.seq, Di.id1, Di.id2, Di.cost,                           
       Pt.id, Pt.geom 
FROM pgr_drivingdistance('SELECT gid::integer AS id,                                       
     Source::integer AS source, Target::integer AS target, 
     Traveltime::double precision AS cost FROM network'::text, 
     2022::bigint, 100000::double precision, false, false) 
   di(seq, id1, id2, cost) 
JOIN netowrk_nodes pt 
ON di.id1 = pt.id;

References: Virtual Box ORACLE VM, OSGeo-Live 11  amd64 iso, Workshop FOSS4G Bonn(​http://workshop.pgrouting.org/2.2.10/en/index.html​),

Since I’ve started working, transport and movement data have been at the core of many of my projects. The spatial nature of movement data makes it interesting for GIScience but typical GIS tools are not a particularly good match.

Dealing with the temporal dynamics of geographic processes is one of the grand challenges for Geographic Information Science. Geographic Information Systems (GIS) and related spatial analysis methods are quite adept at handling spatial dimensions of patterns and processes, but the temporal and coupled space-time attributes of phenomena are difficult to represent and examine with contemporary GIS. (Dr. Paul M. Torrens, Center for Urban Science + Progress, New York University)

It’s still a hot topic right now, as the variety of related publications and events illustrates. For example, just this month, there is an Animove two-week professional training course (18–30 September 2016, Max-Planck Institute for Ornithology, Lake Konstanz) as well as the GIScience 2016 Workshop on Analysis of Movement Data (27 September 2016, Montreal, Canada).

Space-time cubes and animations are classics when it comes to visualizing movement data in GIS. They can be used for some visual analysis but have their limitations, particularly when it comes to working with and trying to understand lots of data. Visualization and analysis of spatio-temporal data in GIS is further complicated by the fact that the temporal information is not standardized in most GIS data formats. (Some notable exceptions of formats that do support time by design are GPX and NetCDF but those aren’t really first-class citizens in current desktop GIS.)

Most commonly, movement data is modeled as points (x,y, and optionally z) with a timestamp, object or tracker id, and potential additional info, such as speed, status, heading, and so on. With this data model, even simple questions like “Find all tracks that start in area A and end in area B” can become a real pain in “vanilla” desktop GIS. Even if the points come with a sequence number, which makes it easy to identify the start point, getting the end point is tricky without some custom code or queries. That’s why I have been storing the points in databases in order to at least have the powers of SQL to deal with the data. Even so, most queries were still painfully complex and performance unsatisfactory.

So I reached out to the Twitterverse asking for pointers towards moving objects database extensions for PostGIS and @bitnerd, @pwramsey, @hruske, and others replied. Amongst other useful tips, they pointed me towards the new temporal support, which ships with PostGIS 2.2. It includes the following neat functions:

  • ST_IsValidTrajectory — Returns true if the geometry is a valid trajectory.
  • ST_ClosestPointOfApproach — Returns the measure at which points interpolated along two lines are closest.
  • ST_DistanceCPA — Returns the distance between closest points of approach in two trajectories.
  • ST_CPAWithin — Returns true if the trajectories’ closest points of approach are within the specified distance.

Instead of  points, these functions expect trajectories that are stored as LinestringM (or LinestringZM) where M is the time dimension. This approach makes many analyses considerably easier to handle. For example, clustering trajectory start and end locations and identifying the most common connections:

animation_clusters

(data credits: GeoLife project)

Overall, it’s an interesting and promising approach but there are still some open questions I’ll have to look into, such as: Is there an efficient way to store additional info for each location along the trajectory (e.g. instantaneous speed or other status)? How well do desktop GIS play with LinestringM data and what’s the overhead of dealing with it?


This post is part of a series. Read more about movement data in GIS.

This post covers a simple approach to calculating isochrones in a public transport network using pgRouting and QGIS.

For this example, I’m using the public transport network of Vienna which is loaded into a pgRouting-enable database as network.publictransport. To create the routable network run:

select pgr_createTopology('network.publictransport', 0.0005, 'geom', 'id');

Note that the tolerance parameter 0.0005 (units are degrees) controls how far link start and end points can be apart and still be considered as the same topological network node.

To create a view with the network nodes run:

create or replace view network.publictransport_nodes as
select id, st_centroid(st_collect(pt)) as geom
from (
	(select source as id, st_startpoint(geom) as pt
	from network.publictransport
	) 
union
	(select target as id, st_endpoint(geom) as pt
	from network.publictransport
	) 
) as foo
group by id;

To calculate isochrones, we need a cost attribute for our network links. To calculate travel times for each link, I used speed averages: 15 km/h for buses and trams and 32km/h for metro lines (similar to data published by the city of Vienna).

alter table network.publictransport add column length_m integer;
update network.publictransport set length_m = st_length(st_transform(geom,31287));

alter table network.publictransport add column traveltime_min double precision;
update network.publictransport set traveltime_min = length_m  / 15000.0 * 60; -- average is 15 km/h
update network.publictransport set traveltime_min = length_m  / 32000.0 * 60 where "LTYP" = '4'; -- average metro is 32 km/h

That’s all the preparations we need. Next, we can already calculate our isochrone data using pgr_drivingdistance, e.g. for network node #1:

create or replace view network.temp as
 SELECT seq, id1 AS node, id2 AS edge, cost, geom
  FROM pgr_drivingdistance(
    'SELECT id, source, target, traveltime_min as cost FROM network.publictransport',
    1, 100000, false, false
  ) as di
  JOIN network.publictransport_nodes pt
  ON di.id1 = pt.id;

The resulting view contains all network nodes which are reachable within 100,000 cost units (which are minutes in our case).

Let’s load the view into QGIS to visualize the isochrones:

isochrone_publictransport_1

The trick is to use data-defined size to calculate the different walking circles around the public transport stops. For example, we can set up 10 minute isochrones which take into account how much time was used to travel by pubic transport and show how far we can get by walking in the time that is left:

1. We want to scale the circle radius to reflect the remaining time left to walk. Therefore, enable Scale diameter in Advanced | Size scale field:

scale_diameter

2. In the Simple marker properties change size units to Map units.
3. Go to data defined properties to set up the dynamic circle size.

datadefined_size

The expression makes sure that only nodes reachable within 10 minutes are displayed. Then it calculates the remaining time (10-"cost") and assumes that we can walk 100 meters per minute which is left. It additionally multiplies by 2 since we are scaling the diameter instead of the radius.

To calculate isochrones for different start nodes, we simply update the definition of the view network.temp.

While this approach certainly has it’s limitations, it’s a good place to start learning how to create isochrones. A better solution should take into account that it takes time to change between different lines. While preparing the network, more care should to be taken to ensure that possible exchange nodes are modeled correctly. Some network links might only be usable in one direction. Not to mention that there are time tables which could be accounted for ;)

This post is a quick instruction for installing Postgres 9.2, PostGIS 2.0 and pgRouting 2.0.

  1. For Postgres, download the installer from enterprisedb.com.
  2. Run the installer. You’ll have to pick a superuser password – remember it, you’ll need it again soon.
  3. At the end of the installation process, allow the installer to start Stack Builder.
  4. In Stack Builder, select the Postgres 9.2 installation and install PostGIS from the list of available extensions.
  5. The PostGIS installation procedure will prompt you for the superuser password you picked before.
  6. I suggest letting the installer create a sample database We’ll need it later anyway.

Now for the pgRouting part:

  1. Download the pgRouting zip file for your system (32 or 64 bit) from Winnie.
  2. Unzip the file. It contains bin, lib and share folders as well as two text files.
  3. Copy these folders and files over to your Postgres installation. In my case C:\Program Files\PostgreSQL\9.2

Installation – done.

Next, fire up pgAdmin. If you allowed the PostGIS installer to create a sample database, you will find it named postgis20 or similar. Otherwise just create a new database using the PostGIS template database. You can enable pgRouting in a database using the following steps:

  1. In postgis20, execute the following query to create the pgrouting extension. This will add the pgRouting-specific functions:
    CREATE EXTENSION pgrouting;
  2. Test if everything worked fine:
    SELECT pgr_version();

    It should return "(2.0.0-dev,v2.0.0-beta,18,a3be38b,develop,1.46.1)" or similar – depending on the version you downloaded.

pgadmin_pgrouting

How about some test data? I’ll be using the public transport network of the city of Vienna provided in GeoJSON format from http://data.wien.gv.at/daten/geo?service=WFS&request=GetFeature&version=1.1.0&typeName=ogdwien:OEFFLINIENOGD&srsName=EPSG:4326&outputFormat=json:

    1. Just copy paste the url in Add Vector Layer | Protocol to load the dataset.
    2. Use DB Manager to load the layer into your database. (As you can see in the screenshot, I created a schema called network but that’s optional.)

import_publictransport

  1. To make the line vector table routable, we use pgr_createTopology. This function assumes the columsn called “source” and “target” exist so we have to create those as well:
    alter table network.publictransport add column source integer;
    alter table network.publictransport add column target integer;
    select pgr_createTopology('network.publictransport', 0.0001, 'geom', 'id');
    

    I’m quite generously using a tolerance of 0.0001 degrees to build the topology. Depending on your dataset, you might want to be more strict here.

  2. Let’s test it! Route from source #1 to target #3000 using pgr_dijkstra:
    SELECT seq, id1 AS node, id2 AS edge, cost, geom
      FROM pgr_dijkstra(
        'SELECT id, source, target, st_length(geom) as cost FROM network.publictransport',
        1, 3000, false, false
      ) as di
      JOIN network.publictransport pt
      ON di.id2 = pt.id ;

    Note how the query joins the routing results and the network table together. (I’m aware that using the link length as a cost attribute will not lead to realistic results in a public transport network but bear with me for this example.)

  3. We can immediately see the routing results using the Load as new layer option:

select_dijkstra
route

Nice work! pgRouting 2.0 has come a long way. In a post from April this year, Boston GIS even announced to add pgRouting into the Stack Builder. That’s going to make the installation even more smooth.

%d bloggers like this: