A utility library for calculations with geolocations.
Features:
- Convert between locations formats (like
[lon, lat]and{lat, lng}) - Calculate distance and heading between locations, move a distance to a new location
- Calculate bounding box around a list with locations
- Calculate whether a location lies inside a circle, bounding box, or polygon
- Calculate the closest point of approach (CPA)
- Uses plain objects, so it plays nice with other libraries, REST API's, and is easy to serialize.
npm install geolocation-utils
import {
toLatLon, toLatitudeLongitude, headingDistanceTo, moveTo, insidePolygon
} from 'geolocation-utils'
// convert various location formats
console.log(toLatLon([4, 51]))
// { lat: 51, lon: 4 }
console.log(toLatitudeLongitude({ lat: 51, lng: 4 }))
// { latitude: 51, longitude: 4 }
// calculate the distance between locations, move to a new location
const location1 = {lat: 51, lon: 4}
const location2 = {lat: 51.001, lon: 4.001 }
console.log(headingDistanceTo(location1, location2))
// {
// heading: 32.182377166258156, // degrees
// distance: 131.52837683622332 // meters
// }
console.log(moveTo(location1, {heading: 32.1, distance: 131.5}))
// {
// lat: 51.00100069207888,
// lon: 4.000997477543457,
// }
// check whether a location is inside a circle, bounding box, or polygon
const polygon = [
[4.03146, 51.9644],
[4.03151, 51.9643],
[4.03048, 51.9627],
[4.04550, 51.9600],
[4.05279, 51.9605],
[4.05215, 51.9619],
[4.04528, 51.9614],
[4.03146, 51.9644]
]
console.log(insidePolygon([4.03324, 51.9632], polygon)) // true
// and much more:
// - calculate bounding boxes around a list of locations,
// - calculate the closest point of approach of two moving objects,
// - etc...| Name | Structure | Description |
|---|---|---|
LatLon |
{lat: number, lon: number} |
lat/lon object |
LatLng |
{lat: number, lng: number} |
lat/lng object |
LatitudeLongitude |
{latitude: number, longitude: number} |
latitude/longitude object |
LonLatTuple |
[longitude: number, latitude: number] |
array with two entries: lon, lat (MIND THE ORDER!) |
Location |
LatLon, LatLng, LatitudeLongitude, or LonLatTuple |
any geolocation structure |
BoundingBox |
{[topLeft: Location, bottomRight: Location]} |
top left and bottom right locations describing a bounding box |
HeadingDistance |
{heading: number, distance: number} |
object containing a property heading in degrees, and distance in meters |
LocationHeadingSpeed |
{location: Location, speed: number, heading: number} |
object containing a location, a heading in degrees, and a speed in meters per second |
TimeDistance |
{time: number, distance: number} |
object containing a time in seconds and a distance in meters |
Create a location object of a specific type.
Available types: 'LonLatTuple', 'LatLon', 'LatLng', 'LatitudeLongitude'.
createLocation(51, 4, 'LatLon') // {lat: 51, lon: 4}Get the latitude of a location
getLatitude({lat: 51, lon: 4}) // 51Get the type of a location object. Returns the type of the location object.
Recognized types: 'LonLatTuple', 'LatLon', 'LatLng', 'LatitudeLongitude'.
getLocationType({lat: 51, lon: 4}) // 'LatLon'Get the longitude of a location
getLongitude({lat: 51, lon: 4}) // 4Test whether two locations are equal or approximately equal. The third argument epsilon is optional and is 0 by default (i.e. only locations that are strictly equal return true).
// compare locations in different formats
isEqual({longitude: 4, latitude: 51}, [4, 51]) // true
isEqual({longitude: 4.1, latitude: 51.2}, [4, 51]) // false
// A calculation like 0.1 + 0.2 introduces a round off error, which can be dealt
// with by passing a value for epsilon specifying the maximum absolute difference.
const epsilon = 1e-12 // max absolute difference
isEqual({longitude: 0, latitude: 0.1 + 0.2}, {lon: 0, lat: 0.3}) // false
isEqual({longitude: 0, latitude: 0.1 + 0.2}, {lon: 0, lat: 0.3}, epsilon) // trueTest whether an object is an object containing numeric properties lat and lon.
isLatLon({lat: 51, lon: 4}) // true
isLatLon({latitude: 51, longitude: 4}) // falseTest whether an object is an object containing numeric properties lat and lng.
isLatLng({lat: 51, lng: 4}) // true
isLatLng({latitude: 51, longitude: 4}) // falseTest whether an object is an object containing numeric properties latitude and longitude.
isLatitudeLongitude({latitude: 51, longitude: 4}) // true
isLatitudeLongitude({lat: 51, lon: 4}) // falseTest whether an object is an array containing two numbers (longitude and latitude).
isLonLatTuple([4, 51]) // true
isLonLatTuple({latitude: 51, longitude: 4}) // falseConvert a location into an object with properties lat and lng.
toLatLng([4, 51]) // {lat: 51, lng: 4}
toLatLng({latitude: 51, longitude: 4}) // {lat: 51, lng: 4}Convert a location into an object with properties lat and lon.
toLatLon([4, 51]) // {lat: 51, lon: 4}
toLatLon({latitude: 51, longitude: 4}) // {lat: 51, lon: 4}Convert a location into an object with properties latitude and longitude
toLatitudeLongitude([4, 51]) // {latitude: 51, longitude: 4}
toLatitudeLongitude({lat: 51, lon: 4}) // {latitude: 51, longitude: 4}Convert a location into a tuple [longitude, latitude], as used in the geojson standard
Note that for example Leaflet uses a tuple
[latitude, longitude]instead, be careful!
toLonLatTuple({latitude: 51, longitude: 4}) // [4, 51]
toLonLatTuple({lat: 51, lon: 4}) // [4, 51]Calculate the average of a list with locations. Returns null when the input is an empty Array or no Array. Returned Location format is the same as the format of the first entry of locations. The array can contain mixed content.
const locations = [
{lat: 51.01, lon: 4.00},
{lat: 51.08, lon: 4.01},
{lat: 51.09, lon: 4.02}
]
average(locations) // {lat: 51.06, lon: 4.01}Calculate the heading and distance between two locations. Returns an object with a property heading in degrees, and a property distance in meters.
headingDistanceTo({lat: 51, lon: 4}, {lat: 51.0006, lon: 4.001})
// {
// distance: 96.7928594737802, // meters
// heading: 46.3657229580657 // degrees
// }Calculate the heading from one location to another location. Returns an heading in degrees.
headingTo({lat: 51, lon: 4}, {lat: 51.0006, lon: 4.001})
// 46.3657229580657 degreesCalculate the distance between two locations. Returns a distance in meters.
distanceTo({lat: 51, lon: 4}, {lat: 51.0006, lon: 4.001})
// 96.7928594737802 metersGet the bounding box of a list with locations, optionally with an extra margin around it in meters. Returns a BoundingBox with topLeft and bottomRight location.
const locations = [
{lat: 50, lon: 10},
{lat: 40, lon: 40},
{lat: 30, lon: 20}
]
getBoundingBox(locations)
// {
// topLeft: {lat: 50, lon: 10},
// bottomRight: {lat: 30, lon: 40}
// }
const margin = 10000 // meters
getBoundingBox(locations, margin)
// {
// topLeft: { lat: 50.08983152841195, lon: 9.860246950846237 },
// bottomRight: { lat: 29.910168471588047, lon: 40.10372851422071 }
// }Test whether a location lies inside a given bounding box.
Returns true when the location is inside the bounding box or on the edge.
The function is resilient against mixing up locations of the bounding boxes' topLeft and bottomRight location.
const boundingBox = {
topLeft: {lat: 20, lon: 0},
bottomRight: {lat: 0, lon: 10}
}
insideBoundingBox({lat: 15, lon: 5}, boundingBox) // true
insideBoundingBox({lat: 20, lon: 5}, boundingBox) // true
insideBoundingBox({lat: 21, lon: 5}, boundingBox) // falseTest whether a location lies inside a circle defined by a center location and a radius in meters.
Returns true when the location is inside the circle or on the edge.
const center = {lat: 51, lon: 4}
const radius = 10000 // meters
insideCircle({lat: 51.03, lon: 4.05}, center, radius) // true
insideCircle({lat: 51.3, lon: 4.5}, center, radius) // falseTest whether a location lies inside a given polygon
Returns true when the location is inside the polygon or on the edge.
// an L shaped area in the port of rotterdam
const polygon = [
[4.031467437744141, 51.96441845630598],
[4.031510353088379, 51.96431268689964],
[4.03048038482666, 51.962779002459634],
[4.045500755310059, 51.96000237127137],
[4.052796363830566, 51.960557711268194],
[4.052152633666992, 51.96198569681285],
[4.045286178588867, 51.96140393041545],
[4.031467437744141, 51.96441845630598]
]
insidePolygon([4.033248424530029, 51.963294643601216], polygon) // true
insidePolygon({lon: 4.033248424530029, lat: 51.963294643601216}, polygon) // true
insidePolygon([4.04545783996582, 51.961668370622995], polygon) // falseMove to a new location from a start location, heading (in degrees), and distance (in meters).
moveTo({lat: 51, lon: 4}, {distance: 100 /*meter*/, heading: 45})
// { lat: 51.000635204829045, lon: 4.00100935046453 }Calculate the closest point of approach (CPA) for two moving objects, for example two ships or cars.
Both tracks contain a location, speed in meters per second, and heading in degrees.
The returned result contains the time and distance of the moment when the two moving objects are closest.
The time in seconds and distance in meters.
// the following two ships are near each other and closing in
const ship1 = {
location: {lon: 4.61039, lat: 51.70401},
speed: 5, // meters/second
heading: 200 // degrees
}
const ship2 = {
location: {lon: 4.60109, lat: 51.69613},
speed: 0.8334, // meters/second
heading: 180 // degrees
}
// lets see how close these ships will get and when they are closest
const { time, distance } = cpa(ship1, ship2)
// time = 251.22255125913932 seconds
// distance = 231.90976012822378 metersNormalize a heading in degrees into the range [0, 360) (lower bound included, upper bound excluded).
normalizeHeading(45) // 45 degrees
normalizeHeading(-40) // 320 degrees
normalizeHeading(380) // 20 degreesNormalize a latitude into the range [-90, 90] (upper and lower bound included). For example a latitude of 91 is normalized into 89 (crossed the pool one degree).
normalizeLatitude(45) // 45 degrees
normalizeLatitude(91) // 89 degrees
normalizeLatitude(-180) // 0 degreesNormalize a longitude into the range (-180, 180] (lower bound excluded, upper bound included).
normalizeLongitude(120) // 120 degrees
normalizeLongitude(200) // -160 degrees
normalizeLongitude(-720) // 0 degreesNormalize the longitude and latitude of a location.
Latitude will be in the range [-90, 90] (upper and lower bound included).
Lontitude will be in the range (-180, 180] (lower bound excluded, upper bound included).
normalizeLocation([360, 95]) // [ 0, 85 ] degreesConvert an angle in degrees into an angle in radians.
degToRad(45) // 0.7853981633974483 radiansConvert an angle in radians into an angle in degrees.
radToDeg(Math.PI / 4) // 45 degrees Convert a speed in knots into a speed in meter per second. 1 knot is 0.514444 m/s.
knotsToMeterPerSecond(10) // 5.14444 m/sConvert a speed in meter per second into a speed in knots. 1 knot is 0.514444 m/s.
meterPerSecondToKnots(10) // 19.438461717893492 knotsConvert a speed in knots into a speed in kilometer per hour. 1 knot is 1.852 kilometer per hour.
knotsToKmPerHour(10) // 18.52 km/hConvert a speed in kilometer per hour into a speed in knots. 1 knot is 1.852 kilometer per hour.
knotsToKmPerHour(10) // 18.52 km/hReturns the earth radius in meters: 6378137.
EARTH_RADIUS // 6378137 metersThe location and orientation of ships and airplaines on the globe is typically expressed in terms of longitude, latitude, and heading.
- Longitude (horizontal axis) is zero degrees at Greenwich, and varies from -180 degrees West to 180 degrees East.
- Latitude (vertical axis) is 0 degrees at the equator, and varies from -90 degrees on the South pool to 90 degrees on the North pool.
- Heading varies from 0 to 360 degrees. North is 0 degrees, East is 90 degrees, South is 180 degrees, and West is 270 degrees.
Source: https://www.e-education.psu.edu/natureofgeoinfo/c2_p11.html
Source: https://aviation.stackexchange.com/a/1597
Yeah, it has bitten you probably too. What is the correct order of a geolocation tuple? Is it [lat, lon] or [lon, lat]? Turns out... there is no right way. There is an ISO standard (ISO 6709), but only half of the big geolocation software libraries adhere to this standard whilst others do the opposite. It's painful and embarrassing having to scratch your head over this out again and again. Like Shane puts it in an answer on StackOverflow:
This situation has wreaked unimaginable havoc on project deadlines and programmer sanity.
A nice overview is written down by Tom Macwright in the blog lon lat lon lat. Here a copy (with some additions):
| Category | [lon, lat] | [lat, lon] |
|---|---|---|
| formats | GeoJSON, KML, Shapefile, WKT, WKB, geobuf | GeoRSS, Encoded Polylines (Google) |
| javascript apis | OpenLayers, d3, ArcGIS API for JavaScript, Mapbox GL JS | Leaflet, Google Maps API |
| mobile apis | Tangram ES | Google Maps iOS/Android, Apple MapKit |
| misc | OSRM, MongoDB, Redis |
The geolocation-utils library supports the [lon, lat] format.
To prevent issues in this regard, it is safest to use an object instead of an array. There are various options:
{latitude: 51.9280573, longitude: 4.4203587} // HTML5 geolocation API
{lat: 51.9280573, lon: 4.4203587}
{lat: 51.9280573, lng: 4.4203587} // Leaflet, MongoDB, and Google Maps
TODO: write out which object stucture are supported by which libraries exactly
To generate ES5 code in the lib folder, run:
npm run build
TODO: create a bundle for browser usage
To test the library (using Ava), run:
npm test
To automatically re-test on change of source code, run:
npm test -- --watch
-
Update version number in
package.json -
Describe version number, date, and changes in
CHANGELOG.md -
Commit the changes, and create a tag with version number
git add ... git commit -m "Released version 1.2.3" git push origin master git tag v1.2.3 git push --tags -
Publish the library
npm publishBefore publishing, the library will be built and the unit tests will run.
MIT