Investigating voice-based co-driver support for EV truck drivers

Abstract

The increasing electrification of heavy-duty freight transport introduces new operational demands for long-haul drivers. Compared with diesel vehicles, electric trucks require continuous balancing of battery levels, charger availability, mandated rest breaks, and changing environmental conditions. These factors increase cognitive load and stress in range-critical moments, while current in-cab systems are predominantly visual and not always practical during driving.

This thesis investigates how voice-based systems can support decision-making in electric trucking by acting as auditory co-drivers. Using a user-centered, Double-Diamond approach, three interaction styles were prototyped and evaluated in video-based simulations: one-way guidance, two-way dialogue, and proactive co-piloting. Data were collected through SUS, SAM, and trust ratings, complemented by think-aloud and interviews.

Findings indicate that proactive, context-sensitive interaction improved perceived trust, clarity, and sense of control, and reduced stress. Timing, clarity of rationale, and autonomy-preserving dialogue emerged as critical design factors. The thesis distils practical design recommendations for voice-based driver support in electric logistics and outlines directions for learnability, adaptive dialogue, and multimodal feedback in high-stakes transport.