This project was concerned with developing an application that takes elevation data and uses it to weight a graph in terms of how energetically demanding travel is between points. It uses existing research on the calculation of energy expenditure for walkers going uphill, downhill and on flat surfaces. The aim is to provide users with paths that will take the least energy to follow, and to give an estimate of calories burned for individual users.
Based on a fork of the Graphhopper Routing Engine - see that project's readme below. I would like to open source this work and contribute it back to the project.
Usage: To use the custom weighting, set the weighting to 'calorie' in the URL.
Guidelines on how to start GraphHopper from source can be found here: https://github.com/graphhopper/graphhopper/blob/master/docs/core/quickstart-from-source.md ; ignore the instructions to clone the repository and simply cd into the graphhopper folder.
To startup the server using your own details, use the following startup command in the command line, replacing the details below with your own:
./graphhopper.sh web greater-london-latest.osm.pbf weight={weight in kilograms} load={load carried in kilograms} height={height in centimetres} female=true age={age in years}
N.B. Height, weight, load and age should be integers. The graph will be weighted according to the details you specify.
Example of a route showing calories required:
GraphHopper is a fast and memory efficient Java routing engine released under Apache License 2.0. Per default it uses OpenStreetMap and GTFS data but can import other data sources.
To get started read through our documentation and install the GraphHopper Web Service locally:
- unstable: docs, zip, Android APK
- 0.9.0 latest stable: announcement, docs, zip, Android APK
- 0.8.2: announcement, docs, zip, Android APK
- 0.7.0: announcement, docs, zip, Android APK
All questions can go to our forum where we also have subsections specicially for developers, mobiles usage (iOS&Android) and our map matching component. Another place to ask questions would be on Stackoverflow but please do not use our issue section. Create new issues only if you are sure that this is a bug and see how to contribute in the next section.
See GraphHopper in action on GraphHopper Maps
GraphHopper Maps uses the Directions API for Business under the hood, which provides a Routing API via GraphHopper, a Route Optimization API via jsprit, a fast Matrix API and an address search via Photon. Additionally the map tiles from various providers are used where the default is Omniscale, and all is available for free, via encrypted connections and from German servers for a nice and private route planning experience!
There are subprojects to make GraphHopper work offline on Android and iOS
GraphHopper supports several routing algorithms like
Dijkstra and
A* and its bidirectional variants.
Furthermore it allows you to use
Contraction Hierarchies (CH)
very easily, we call this speed mode and without this CH preparation we call it flexible mode.
The speed mode comes with very fast and lightweight (less RAM) responses and that although it does not use heuristics in its default settings. The downsides are that the speed mode allows only pre-defined vehicle profiles (multiple possible in GraphHopper) and requires a time consuming and resource intense preparation. And implementing certain features are not possible or very complex compared to the flexible mode.
The hybrid mode also requires preparation time and memory, but is much more flexible regarding changing properties per request or e.g. integrating traffic data and more. Furthermore this hybrid mode is slower than the speed mode but it is an order of magnitude faster than the flexible mode and uses also less RAM for one request.
You can switch between all modes at request time.
We chose the Apache License to make it easy for you to embed GraphHopper in your products, even closed source. We suggest to contribute back your changes, as GraphHopper evolves fast, but of course this is not necessary.
OpenStreetMap is directly supported from GraphHopper. Without the amazing data from OpenStreetMap GraphHopper wouldn't be possible at all. Other map data will need a custom import procedure, see e.g. Ordnance Survey, Shapefile like ESRI or Navteq.
GraphHopper is written in Java and runs on Linux, Mac OS X, Windows, BSD, Solaris, Raspberry Pi, Android, Blackberry and even iOS.
Embed GraphHopper with OpenStreetMap support into your Java application via the following snippet
<dependency>
<groupId>com.graphhopper</groupId>
<artifactId>graphhopper-reader-osm</artifactId>
<version>[LATEST-VERSION]/version>
</dependency>If you want to write your own import procedure or you don't need OSM import like on Android, then use:
<dependency>
<groupId>com.graphhopper</groupId>
<artifactId>graphhopper-core</artifactId>
<version>[LATEST-VERSION]</version>
</dependency>We've build the GraphHopper class which makes simple things easy and complex things like multi-modal routing possible. Still you can use the low level API of GraphHopper and you'll see that it was created to allow fast and memory efficient use of the underlying datastructures and algorithms.
On Android and Blackberry (since 10.2.1) we provide an integration with Mapsforge which makes offline navigation one step closer. Due to the usage of memory mapped files and Contraction Hierarchies we avoid allocating too much memory which makes it possible to run Germany-wide queries with only 32MB in a few seconds. We provide an Android studio project as well as the Maven-Android integration to be used in other IDEs.
With the web module we provide code to query GraphHopper over HTTP and decrease bandwidth usage as much as possible.
For that we use a polyline encoding from Google, the Ramer–Douglas–Peucker algorithm and a simple
GZIP servlet filter.
On the client side we provide Java and JavaScript code (via Leaflet) to consume that service and
visualize the routes.
GraphHopper also runs on the Desktop in a Java application without internet access. E.g. you could use the rough user interface called MiniGraphUI provided in the tools module, see some visualizations done with it here. A fast and production ready map visualization for the Desktop can be easily implemented via mapsforge.
Install GraphHopper via Docker. You only need to change the docker-compose.yml entrypoint in core/files/ to run whatever map you like. Then just type:
cd core/files/
docker-compose up -dHere is a list of the more detailed features including a link to the documentation:
- Simple start for users - just Java necessary! Simple start for developers due to Maven.
- Works out of the box with OpenStreetMap (osm/xml and pbf) but can be adapted to use your own data
- OpenStreetMap integration: Takes care of the road type, the surface, barriers, access restrictions, ferries, conditional access restrictions, ...
- GraphHopper is fast. And with the so called "Contraction Hierarchies" it can be even faster (enabled by default).
- Memory efficient data structures, algorithms and the low and high level API is tuned towards ease of use and efficiency
- Provides a simple web API including JavaScript and Java clients
- Multiple weightings (fastest/shortest/...) and pre-built routing profiles: car, bike, racingbike, mountain bike, foot, motorcycle, ...
- Offers turn instructions in more than 35 languages, contribute or improve here
- Displays and takes into account elevation data (per default disabled)
- Can apply real time changes to edge weights (flexible and hybrid mode only)
- Customize vehicle profiles per request (flexible and hybrid mode only)
- Possibility to specify a heading parameter of the vehicle for start, end and via points for navigation applications via
pass_throughorheadingparameters (flexible and hybrid mode only) - Alternative routes (flexible and hybrid mode only)
- Turn costs and restrictions (flexible and hybrid mode only)
- Country specific routing via SpatialRules
- Multiple profiles and weightings
- Several pre-built routing profiles: car, bike, racingbike, mountain bike, foot, motorcycle, ...
- The core uses only a few dependencies (hppc, jts and slf4j)
- Scales from small indoor-sized to world-wide-sized graphs
- Find nearest point on street e.g. to get elevation or 'snapp to road'
- Do map matching with GraphHopper