Your browser doesn't support javascript. This means that the content or functionality of our website will be limited or unavailable. If you need more information about Vinnova, please contact us.

Impairment-Aware Intelligent Mobility System (I-AIMS)

Reference number
Coordinator Göteborgs universitet - Göteborgs universitet Inst f Tillämpad IT
Funding from Vinnova SEK 995 630
Project duration November 2023 - November 2024
Status Completed
Venture Behavioral changes in the mobility system
Call Human-centered mobility - The mobility system of the future based on human behavior and perception

Important results from the project

The I-AIMS project investigated how to utilize Smart-Eye’s Driver Monitoring System (DMS) with a large language model (LLM) to regulate impaired driver states. We found LLMs can be beneficial to driver state but must be used in context sensitive ways in relation to safety critical scenarios: Type of feedback, type of driving autonomy, and scenario parameters (visibility, proximity of safety critical event) all differently affect driver states and perceived relevance/acceptance of the feedback.

Expected long term effects

The significance to drivers of being able to use an LLM assistant would be to provide easy access to information about upcoming events of a safety critical nature and can also serve to alleviate negative affective states that may contribute to safety critical events. Furthermore, in drivers less experienced with modern in cabin technology, LLM assistants may provide assurance, reducing stress, and thereby reducing potential for negative outcomes/involvement in safety critical scenarios.

Approach and implementation

The project consisted of several phases: 1) development of LLM prototype; 2) EPM application for conducting a pilot study using human participants; 3) pre-pilot study for participants to rate aspects of LLM feedback; 4) development of simulated scenarios using the CARLA simulation tool; 5) study (3 conditions : control, manual-drive, autonomous drive) using 30 participants to evaluate the LLM feedback on driver state in simulated driving scenarios. 6) analysis/submission for 2 conference pubs.

The project description has been provided by the project members themselves and the text has not been looked at by our editors.

Last updated 16 December 2024

Reference number 2023-03068