The BMW Group is working on a parking prediction system designed to shorten the search for vacant on-street parking, particularly in cities. Together with its partner INRIX, a leading provider of transportation intelligence and connected car services, BMW is presenting a research prototype of the Dynamic Parking Prediction application at TU-Automotive Detroit. The system is being displayed in a BMW i3.
Dynamic Parking Prediction is able to predict parking availability using on-road vehicle movement data. In this way the application is able to shorten the search for vacant on-street parking, particularly in cities, and provides an effective way of reducing parking-related traffic. Initial tests with this prototype on an i3 have already been successfully completed in Munich. The system is self-teaching and can therefore easily be rolled out to other cities too.
The BMW Group has been researching solutions to take the stress out of parking and to reduce the time spent locating a vacant space since 2011. For the purposes of this project, up-to-date digital maps were produced showing all public parking spaces, while several thousand vehicles from a test fleet supplied anonymous movement data generated when using these spaces.
Data was supplied by fleet vehicles both when leaving a parking space and also when searching for a space. Based on the digital map, the local prediction algorithm and the parking data from the fleet vehicles, the research application calculates current parking options in a given area—e.g., a particular part of town.
This information is then presented on the dashboard display. The number of currently vacant parking spaces and the number of drivers looking for parking are both factored into the calculation. Even when the system is restricted to using data just from the fleet vehicles it achieves reliable results; prediction accuracy increases in step with the number of vehicles supplying data.
(Ford, which is also tackling the parking problem in its Smart Mobility portfolio, is taking a different approach with its Parking Spotter experiment, conducted with Georgia Tech. This system leverages driver-assist sensors that most Ford vehicles already have, including sonar and radar. The sensors search for open parking spaces while the driver looks for spots around the city, and share the information with a cloud database other drivers can access. Earlier post.)
With the DriveNow fleet vehicles the BMW Group is collecting further useful experience. This parking information service could potentially be rolled out to all other vehicles in the car-sharing fleet in the near future.
INRIX and BMW will pool their expertise to further refine the research prototype for use in production vehicles.
BMW was early to understand the importance of connectivity between cars, drivers and their surroundings, and with BMW ConnectedDrive it has gone on to become a leader in this field. The company is also expanding its position as a premium mobility provider, with intelligent services and applications such as ChargeNow, ParkNow or intermodal navigation.
One major sub-cluster of connected vehicle technology from BMW is Connected Navigation, in which the RTTI Real-Time Traffic Information system is already helping drivers cope more effectively, more safely and more conveniently with today’s driving challenges by providing them with accurate early warning of congestion and hold-ups and informing them of alternative routes.