„How does the increasing levels of driving automation change social interaction in semi-automated road traffic, and how does the integration of sensor-driven media open up new possibilities to understand, design and evaluate such interactions? The unique focus of our approach is that interaction is viewed as a cooperative reality between all road users, cars, sensors, and their environments.“

Executive Summary

Road traffic is a social situation involving the intensive intercommunication and interaction of road users. The behavior of drivers therefore affects not only themselves, but also other road users, such as drivers, cyclists, or pedestrians. With the increasing automation of driving and the delegation of driving tasks to vehicles, the interactions between road users will also see a fundamental transformation: These interactions will be generated cooperatively between vehicles, environments, road users, and the technologies and practices involved.

Due to the gradual automation of driving, roads in the near future will consist of a blend of non-automated, semi-automated, and highly automated vehicles with various sensor technologies and capabilities. Depending on the level of automation of the vehicles, the level of engagement of their respective drivers in communicating with other road users will vary. In other words: Humans will increasingly have to communicate with machines. This requires new forms of interaction between road users in mixed traffic situations (e.g. (non-)automated vehicles, pedestrians, or cyclists). In addition to the automation of driving, advances in in-vehicle and wearable sensor technologies have opened up the possibility of ad-hoc communication links between vehicles and other road users (vehicle-to-anything). For this reason, communication and interaction between automated vehicles and other road users through the use of new sensory technologies has increasingly become the focus of the field human computer interaction. However, the approaches developed so far are mainly concerned with the effectiveness of communication between road users (i.e. understanding and communicating intentions).

Less is known about the social consequences of such forms of interaction. Research has thus far failed to address the questions of how interaction can be accomplished in a cooperative and situated way, how it reconfigures the interplay between the human and the technical sensorium, and how its specific situatedness can also be considered in the confusing traffic situation “in the wild.”

This project therefore investigates how the increasing automation of driving is changing social interaction in road traffic and how the integration of sensory media opens up new possibilities for understanding, shaping, and assessing such interactions. We will investigate the integration of multimodal sensing, including in-vehicle sensors and wearables, in the design and assessment of highly automated traffic situations in the wild with different road users and a wide range of automation levels. How are traffic situations perceived by drivers, vehicles, and other road users, and how can sensors be used to shape the interaction and negotiation of complex traffic situations? The project focuses in particular on the heterogeneous evaluation criteria of interactions in road traffic and aims to identify behaviors in traffic that are perceived as “prosocial” and promote the well-being of all road users in order to achieve safe, harmonious traffic.

The proposed subproject is situated at the intersection of ubiquitous computing and human-machine interaction research and is pursuing its key project objective by means of a multi-level research design: Firstly, we will use data collected with multimodal sensor technology to identify elements that define behavior in mixed traffic scenarios as prosocial. In the process, we will use cooperation between human cognition and technical sensor technology at a methodological level. Secondly, we will explore new forms of communication between road users based on sensor technologies by identifying what information is needed in what situations and how technologies can be designed in an inclusive and accessible way. Third, the subproject will define interaction concepts for mediation and cooperation between humans and automated vehicles, including levels of a) communication between road users, b) assessment of the reliability of processed information, c) discrimination between cases of evasion and falsification, and d) compliance with individual and social norms and values. Finally, a test environment will be developed along with associated methods in order to investigate social interaction in heterogeneous traffic scenarios by employing quantitative and qualitative methods in a series of rigorous laboratory and field studies. Based on the previous results of the conducted studies, the subproject will ultimately provide a guideline for the development of interactive technologies for prosocial behavior in road traffic.