Dialog with trucks: Situation-aware conversational agents in driver support systems

Purpose and goal:

Conversational Agents (CA) have the potential to provide improved real-time support to drivers. However, current CA disregard situation awareness, a deficiency that can increase the driver’s cognitive load. The goal is to integrate situation-aware CA into driver support systems (SAE L1-2). Our focus is on establishing a smooth conversation flow between long-distance truck drivers and CA and studying the impact in terms of safety, efficiency and user experience.

Expected results and effects:

Expected results: a) A decision system that combines various data to achieve situational awareness and controls CA dialogue based on the situation, b) Understanding of the CA-role for long-distance drivers, c) A CA prototype tailored to the needs of long-distance drivers, d) Initial guidance for the design of CA for optimal driver support. The results will support future R&D among project partners. Dissemination will occur through articles, demonstrations, and an external reference group.

Approach and implementation:

The project consortium consists of Scania (TRATON), Nagoon, KTH, and Smart Eye. The project will develop the following functionalities: Situational awareness, Dialogue flow management, and Safety and ethics policy for CA. A series of experiments will be conducted, each designed to highlight a specific aspect of situational awareness, dialogue flow, and their effects on truck drivers. The investigated factors will be anchored in interviews, observations, and measurements with truck drivers.