*There is a [newer version](https://developers.google.com/maps/navigate-with-car-2024/index) of this document
with examples of how to use new Places API and Routes API.*

Typically when the user is driving to a point-of-interest (POI) they would
actually require navigation to a nearby location such as a parking lot or an
access point. Picture a driver that needs to go to an airport or a shopping
mall. When going to an airport the driver is trying to reach a terminal or a
drop-off point rather than the center of the airport compound. Likewise a
parking lot next to the shopping center might be a much better option for a user
with a car. In this case when developing an app or a service to guide the
driver, first find the location they are going to and then offer the driver
optional destinations near that location. This approach allows you to provide
better service to your customers and demonstrate that you value their time, with
a small but smart development effort that leverages location data provided by
Google Maps Platform.
**\[System Context\] High level navigation systems**   
Simplified view of systems involved when guiding the driver to a location with a car ![High level systems diagram](https://developers.google.com/maps/architecture/images/system_diagram_nav_with_car.png)

We will show how you can use Places API and Directions API in sequence or
combination to identify the right location to park next to your destination.
This solution aims to remove uncertainty and multiple interactions to find a
spot to park while driving once closer to the end destination. Let's look at two
examples of how these cases could be implemented in detail.

## Example 1 - a tourist attraction with no parking lot

Let's pick a location that is not directly connected to the roadside so that a
route must be retrieved separately, the **destination** is **Notre Dame** in
Paris and as a **starting point** let's use the **Gare de l'Est** railway
station (48.87697775149635, 2.3592247806755564).

<br />

![Gare de l'Est](https://developers.google.com/static/maps/architecture/navigate-with-car/images/gare_de_lest.png "Gare de l'Est")

<br />

The driver leaves the station about 16 minutes north of Notre Dame. Notre Dame
is a famous tourist attraction on an island with bridge access, some one-way
streets and no large parking lot available nearby, so this should be an
interesting challenge.

> [!IMPORTANT]
> **Try now:** Visit cloud console and enable the [places API](https://console.cloud.google.com/marketplace/product/google/places.googleapis.com?utm_source=architecture_poi).

### Places API request

This example Places API request is using "text search" to find Notre Dame. See
the [developer
documentation](https://developers.google.com/maps/documentation/places/web-service/search-text)
for details. The "YOUR_KEY" in these examples is the API key you have for the
Google Maps Platform Places and Directions APIs enabled in the Google Cloud
Console. For a more detailed explanation please see the documentation for
[getting started with Google Maps
Platform](https://developers.google.com/maps/get-started).

    https://maps.googleapis.com/maps/api/place/textsearch/json?query=notre_dame&location=48.864716%2C2.349014&region=fr&key=YOUR_KEY

The response to the above request gives the following latitude \& longitude
coordinates:

```
  "results" : [
      {
        "formatted_address" : "Notre Dame, Paris",
        "geometry" : {
           "location" : {
               "lat" : 48.8527288,
                },
  ...
```

As you can see, the coordinates indeed pinpoint Notre Dame.

|---|---|
| ![Notre Dame photo](https://developers.google.com/static/maps/architecture/navigate-with-car/images/notre_dame_photo.png "Notre Dame photo") | ![Notre Dame on map](https://developers.google.com/static/maps/architecture/navigate-with-car/images/notre_dame_map.png "Notre Dame on map") |

As a developer or a service provider, it's a good idea to automatically prompt
the driver "Looking for parking near Notre Dame?". The user experience will be
different depending on the device and screen, but a subtle popup text that
disappears after a short time could be worth considering in this case. To find
parking near Notre Dame, do a Places API text search with type "parking" and
radius "300" parameters set. This example will bias the results to parking areas
within 300 meters of Notre Dame.

    https://maps.googleapis.com/maps/api/place/textsearch/json?query=parking
    &location=48.8527288%2C2.3505635&region=fr&type=parking&radius=300&key=YOUR_KEY

The first result is "Parking Saemes Maubert-Lagrange" at 48.850591, 2.3486436.
Let's use this location in the next part where we are doing a Directions API
request.

### Directions API request

To find the route from the station to a parking area close to Notre Dame, you
need to do a Directions API request. Just set the origin and destination
parameters. See [developer
documentation](https://developers.google.com/maps/documentation/directions/get-directions)
for more options.

    https://maps.googleapis.com/maps/api/directions/json?origin=48.8767903,2.3592251&destination=48.850591%2C2.3486436&key=YOUR_KEY

Direction API response typically contains multiple route suggestions. Each route
consists of multiple "legs" and those contain multiple "steps" showing traveling
distance and coordinates. To find the final coordinates reachable with a car,
use the "end_location" field of the last step of a last leg in a route you
selected.

This location may seem like not the first choice, but when you look at the map
or Google Street View, you will find that this location is within a very short
walking distance, about 6-7 minutes, from Notre Dame.

<br />

![Walking route from parking to Notre Dame](https://developers.google.com/static/maps/architecture/navigate-with-car/images/parking_to_notredame.png "Walking route from parking to Notre Dame")

<br />

Walking route from parking area to Notre Dame

## Summary for example 1

As you can see, finding a parking area for a car requires only a few API calls
and displaying the suggestion to the driver. This saves time for your users as
they are not driving around on one-way streets or trying to find a place to
leave the car in a crowded tourist attraction. As a summary we did the following
API requests:

- A Places API request to find Notre Dame coordinates
- A Places API request to find a parking area within 300 meters of those coordinates
- A Directions API request to find a route to the parking area

## Example 2 - Guiding to a terminal in a large airport

<br />

![Heathrow airport on map](https://developers.google.com/static/maps/architecture/navigate-with-car/images/heathrow_map.png "Heathrow airport") Let's consider another example where the driver is trying to reach a large area that has multiple locations, POIs or potential drop-off points. The driver is leaving from Wembley stadium, UK (51.557263604707224, -0.2794575145680608) and trying to get to **London Heathrow airport terminal 5** . Typically the user would start typing in "Heathrow" and most of the time people pick one of the first 2 or 3 results if it looks like the correct result. From the selected result you can find out that the location is an airport by reading the "types" field values. You can find the complete list of types [here](https://developers.google.com/maps/documentation/places/web-service/supported_types).

<br />

      "types" : [ "airport", "point_of_interest", "establishment" ],

However, anyone who has visited any large airport knows that it takes a
significant amount of time to transfer between terminals if you happen to arrive
at the wrong location. In this case, since the type of the search result is
"airport", it is useful to automatically follow up query to search for
"terminal" using the location of the London heathrow airport (51.47016927594547,
-0.45432767852489075). In the example you can also see the optional limitation
of search radius of 3000 meters.

      https://maps.googleapis.com/maps/api/place/textsearch/json?query=terminal&location=51.47016927594547%2C-0.45432767852489075&region=gb&key=YOUR_KEY&radius=3000

The query above gives us a list of the terminals and individual drop-off points
at those terminals. From the list of results we can find "London Heathrow
Airport Terminal 5" and its place id:

     "place_id" : "ChIJtQRd6XVxdkgRTUGZtcsoGNc",

Using the "place_id" parameter we can tell the Directions API that we are
trying to reach a place rather than some arbitrary coordinates.

      https://maps.googleapis.com/maps/api/directions/json?origin=51.557263604707224,-0.2794575145680608&destination=place_id:ChIJtQRd6XVxdkgRTUGZtcsoGNc&key=YOUR_KEY

Again, like in the previous example, by reading all the steps in the JSON
response you will find the last part of the journey and the end locations lat \&
long coordinates (51.47079979999999,-0.4896765).
![Heathrow terminal 5](https://developers.google.com/static/maps/architecture/navigate-with-car/images/heathrow_terminal5_streetview.png "Heathrow terminal 5")

<br />

## Summary for example 2

Like in the previous example, the development effort to provide improved
guidance to the driver is very small and the benefits in time saved to the
driver are easily measured. We did the following requests:

- A Places API request to find Heathrow airport
- A Places API request to find all terminals close to Heathrow airport coordinates
- A Directions API request to get a route to terminal 5

## Conclusion

As you can see, the development effort required to provide a premium service to
your customers traveling by car is rather small and allows you to easily
demonstrate how you are thinking about your customers' needs.

For development purposes you can use many of the libraries and tools available
to help you build the services for your customers. Have a look at our publicly
available [GitHub repository for Google Maps
Platform](https://github.com/googlemaps/).

## Next Actions

- [Outdoor to indoor navigation](https://developers.google.com/maps/architecture/outdoor-to-indoor-navigation-with-oriient-and-gmp.md.txt)
- [Building location validation capability with Google maps](https://developers.google.com/maps/architecture/geocoding-address-validation.md.txt)

## Principal author

Mikko Toivanen \| Google Maps Platform Solutions Engineer