HERE AND NOW.
HERE AND NOW.
Mapping specialist HERE, co-owned by AUDI AG, creates the digital map needed to navigate with centimeter precision: the HERE HD Live Map.
By: Johannes Köbler
HERE has enormous experience in the navigation map business—but the company takes a huge leap with the HERE HD Live Map. The new data platform depicts the traffic space as a three-dimensional model with a new level of precision: down to centimeters, instead of the meter standard in GPS, and dynamic instead of static. The HERE HD Live Map forms the digital basis for the piloted driving of the future.
The new map is divided into square tiles, known as map tiles, with each edge measuring 2.0 kilometers (1.243 miles). This division means that immense data volumes can be split into manageable portions. The updates within each tile can be measured in kilobytes. The data is hosted on a HERE backend in the cloud, with transfer to and from the cars most likely handled via the cell phone network. Today’s LTE® standard is already facilitating all sorts of things, while future 5G networks promise to raise data transfer rates and connection speeds even more substantially.
The HERE HD Live Map is structured in three layers. The first of them, the HD map, incorporates a static digital image of the surroundings. Guardrails, stoplights, road signs, curbsides and similar fixed objects form the reference points on which the piloted cars of the future could be able to orient themselves with centimeter precision. The HD map also includes a database of information on hotels, businesses and restaurants. The second layer carries the “live roads” information: a dynamic level that holds current data in nearly real time. Construction zones, accidents, emergencies, black ice or fog—the live roads content comes from different sources but primarily from the sensors of participating cars. The intelligent swarm generates extensive traffic knowledge that is up-to-date on things like “green waves” in city traffic (which sync green lights for reduced congestion), changing speed limits and available parking spaces.
A consortium made up of AUDI AG, the BMW Group and Daimler AG acquired Berlin-based map database HERE from the Nokia Corporation at the end of 2015. Eighty percent of all cars currently driving in Europe and North America with integrated navigation systems are equipped with HERE maps.
The map service remains an open, independent, continuously expanded and updated platform for cloud-based maps and mobility services even after the acquisition. Accessible for all customers within the automotive industry and other sectors, HERE is laying the foundation for piloted driving and the development of new mobility services.
The “humanized driving” section is an investment in the near future. The piloted cars of the future will find out from the HERE data pool how the driver has behaved in a certain situation similar to the current one—whether, for instance, he braked sooner or a little later in response to a slower-moving vehicle on the highway. This monitoring will help enable the system to adapt its behavior to the habits and expectations of the driver.
The HERE HD Live Map is still under development, but some of its functions are already in use by a large number of networked cars in North America and Western Europe. For Audi customers, the map makes demanding driver assistance functions like Traffic jam assist or predictive efficiency assist even more exact and powerful.
In the live roads layer, the HERE map incorporates real-time information on the traffic environment. Together with the HD map, they will be able to help provide a wide range of useful services to car drivers in the near future. Let’s face it: All sorts of things can happen on the daily commute.
Our driver wants to travel from his home to work in piloted mode. There has been an accident involving injuries at an intersection on his route. The paramedic vehicle and police car continuously transmit their location data to the HERE backend. Every user nearby receives a warning tailor-made for him, enabling him to take the necessary avoiding action.
Stoplight information is one of the services provided on the HERE map, providing information to the networked cars and their drivers on the switching phases for the stoplights along their route.
A particular part of the city is currently experiencing a heavy downpour. The cars driving through here identify the situation via their sensors and windshield wipers and notify the backend in the cloud. When a piloted Audi receives the advisory, it reduces its speed in plenty of time. Convertibles and roadsters close their roofs.
The workplace destination is in the city, where there are no employee parking spaces. Our piloted Audi has to drive itself to one of the nearby parking garages, where the centimeter-precise HD map helps it find its way around with great accuracy. In a subsequent HERE iteration, it could receive detailed information on available parking spaces.
The new HERE HD map uses around 80,000 different sources worldwide, including data from traffic control centers and land registries, images from satellites and aircraft, and camera images from fleet vehicles. The lion’s share comes from around 200 HERE vehicles that drive along roads and capture information. Mounted on the roof of the current measurement cars is a rotating LiDAR device (Light Detection And Ranging). It continuously sends ultra-short light pulses on a wavelength of 905 nanometers, invisible to the human eye. The object being scanned reflects the beams back to the receiver unit. The software calculates the distance between object and sensor based on the elapsed time.
The 360-degree scans made by HERE vehicles generate 700,000 pixels per second, adding up to 140 gigabytes of data per day. The pixel cloud creates a 360-degree image of the surroundings extending for a distance of up to 80 meters and a height of 30 meters.
The rolling and nodding movements made by the car while driving are recorded separately and calculated out of the data set. LiDAR technology has considerable strengths but also certain handicaps. Its high angular resolution means it can usually identify objects with great certainty; e.g., in most circumstances it can differentiate between a trash can and a pedestrian at the side of the road, even if they are right next to each other. Because LiDAR systems use their own light, they can also assist the driver at night. However, their performance deteriorates in bad weather such as rain, snow and fog. LiDAR devices do not deliver color or brightness values for their surroundings, instead measuring the intensity of the reflected beam. They can have problems with strong or weak reflections (due to dirt) from traffic signs and road markings. During image generation, this information is processed with the aid of camera data before being integrated into the HD map.