The EMMI project (2020-2023), funded by the German Federal Ministry for Economic Affairs and Energy, investigates solutions for empathic human-machine interfaces for autonomous driving cars.
A successful introduction to the market of partially, highly, and fully automated driving functions requires the acceptance of the technology. Key features include: 1) a high level of confidence in the system’s safety and 2) an individually perceived benefit of this technology. Therefore, EMMI investigates the systematic design, the development, and the evaluation of an empathic human-machine interface with an interactive virtual agent as the primary communication interface for the user.
The ACG’s role in EMMI is to model trustworthy behavior of the virtual agent. This requires an agent that can evaluate situations from his and the user’s perspective (cf. Theory of Mind). Therefore, the ACG will simulate the internal user emotions regarding trust as well as the relationship between user and agent. Hereby, a central goal is to analyze social signals from the users and correlate them with their internal experiences. A first step is to identify and analyze situations that might elicit fear due to a loss of autonomy in critical situations. In a second step, we will transfer concepts known to increase trust in human-human interactions, e.g., social mimicry, and evaluate them in the human-agent-interaction.