SP 5Automated driving through urban junctions

In urban traffic, junctions such as roundabouts and intersections represent key black spots for accidents. Regulating the traffic and road users, reconciling their competing interests, and rapidly changing situations, place exacting demands on the driver’s attentiveness. This means that there is a potentially higher level of danger for road users. Automated driving functions can help to achieve a significant increase in safety in situations like these.

In comparison to human drivers, sensors and functions in an automated vehicle are not susceptible to symptoms of fatigue or potential distractions, and generally make optimal decisions determined by sensing their environment. Moreover, vehicles with suitable sensors are, in contrast to human drivers, consistently able to register all aspects of their surroundings simultaneously and this, in many cases, with a much higher level of precision. The simultaneous evaluation of digital map data also makes localisation easier in complex intersection and roundabout topologies, therefore allowing the predictive algorithms to navigate safely through situations like these.

Intersections and roundabouts

Highly automated driving functions in an urban environment are presented with the challenge of having to safely and reliably master highly varied situations and scenarios. Junctions such as intersections and roundabouts, in particular, do not exhibit a uniform appearance, therefore requiring vehicle reactions that are extremely different depending on the traffic situation. Many different features of the static infrastructure of the junction and the current traffic situation must be factored in before the overall traffic situation can be registered. Both an intersection and a roundabout may also demand a cooperative strategy from automated driving, one which also factors irregular behaviour by other road users.

The ‘Automated driving through urban junctions’ subproject aims to implement safe, automated vehicle guidance through selected intersections and roundabouts. Development will proceed in three stages that will become successively more challenging:

  • To start with, driving straight ahead through intersections will be made possible in a way that factors in all rules governing the right of way, traffic lights, and any crossing traffic. The vehicle will be controlled using the road markings on the starting and destination lanes.
  • Longitudinal traffic, along with road users in the destination lanes, including pedestrians and cyclists crossing the lane, must be factored in when the vehicle is turning right. The vehicle is controlled in this situation by guiding it along a trajectory, planned within the boundaries of the lane.
  • As a final step, oncoming traffic must be factored in for turning left in order to guide the vehicle along a trajectory that crosses the intersection to reach the destination lane.

Detecting and factoring in all relevant road users is also important for the ‘roundabout’ junction. Its development will also proceed over three stages. The first step involves factoring in the relevant road users when approaching the roundabout, with the next step involving driving into the roundabout and through it to reach the required exit point. The last step will factor in driving out of the roundabout.



SP 1

Sensing the environment ­and situational understanding


SP 2

Digital map and localisation


SP 3

Concepts and pilot applications


SP 4



SP 5

Automated driving through urban junctions


SP 6

Automated driving on urban streets


SP 7

Interaction with vulnerable road users