Automated vehicles are no longer a vision of the future – they are on the verge of becoming an everyday part of our road traffic. But how will these vehicles communicate with pedestrians when there is no longer a driver behind the wheel? In spring 2023, the German Aerospace Center (DLR) conducted a study on the interplay between vehicle size, light signals via external human-machine interfaces (eHMIs) and vehicle kinematics and the influence on the experience and behaviour of people on foot. Among other things, the study analysed how the objective and subjectively perceived safety of pedestrians changes under different conditions.
How do vehicles communicate without a driver?
When a car is travelling without a driver, the usual non-verbal communication is no longer possible. For example, a brief eye contact or a nod that signals: "You can go." This is where eHMIs come into play, using light signals to communicate, for example, the automation status. These are intended to indicate to pedestrians that the vehicle is an automated vehicle or, for example, the intention of the vehicle to stop. The DLR study investigated how such eHMIs need to be designed in order to promote trust and safety – especially in vehicles of different sizes, from small cars to buses.
Virtual reality: a test under realistic conditions
In a virtual traffic scenario, test subjects were able to try out how they interact with automated vehicles in an immersive environment. VR goggles were used to immerse the participants in the test environment (see Figure 1). The aim of the experiment was to find out how the participants reacted to different circumstances in the situation (when they decided to cross the road) and how safe they felt in each case. They were faced with the task of deciding when they could safely cross the road in the direction of their destination "underground station" when an automated vehicle approached from the left under different conditions. The test subjects used a hand controller to indicate when they would cross the road.
Figure 1: Bird's eye view of the test environment (test subjects saw the situation from an egocentric perspective).
The vehicles used were small cars and buses that displayed various light signals (eHMI) when they braked:
Static eHMI: A permanently illuminated light strip mounted below the windscreen. The illuminated light strip signalled that the vehicle was in automated mode (see Fig. 2). Dynamic eHMI: A pulsating light that indicates the vehicle's intention to give way. No signal: The vehicle only communicates through its driving behaviour.
Figure 2: Light signals on the vehicles in the test.
Dynamic eHMIs promote trust and safety When vehicles use pulsating light signals to communicate their intentions, pedestrians feel safer and start crossing the road sooner. This applies to both vehicle sizes.
Larger vehicles appear more threatening Buses were perceived as more dangerous. Pedestrians showed higher confidence in large vehicles with static eHMI than without eHMI, while they felt more confident in cars without eHMI compared to static eHMI.
Mismatch leads to uncertainty For example, if a vehicle indicates that it is going to stop but continues on its way, pedestrians feel unsafe. This mismatch was rated particularly negatively.
Why are the results important?
The introduction of highly automated vehicles requires clear and understandable communication methods. Especially in cities, where pedestrians will interact with automated vehicles on a daily basis, solutions such as dynamic eHMIs and coordination with vehicle kinematics are crucial for a safe and efficient traffic environment and are dependent on their size.
What comes next?
The results of the VR study provide important insights for the iterative development of interaction concepts for automated vehicles. Based on these results, DLR plans to investigate eHMIs in more complex, realistic scenarios with several road users and automated vehicles.
The aim is to further investigate safety and confidence in automated vehicles through real vehicle studies with DLR research vehicles and our own eHMI prototypes, thus paving the way for safe, shared mobility in the future.
