I-TSA Traffic Safety Assessment in a Simulator Experiment with Integrated Information and Assistance Systems

Abstract

Summary:The increasing number and complexity of in-vehicle information systems (IVIS) and advanced driver assistance systems (ADAS) require an accurate and timely assessment of their impact on traffic safety even during the development process. The I-TSA evaluation tool, developed within the German research consortium INVENT, offers a standardized procedure for the assessment of traffic safety based on the driving error occurrence in up to 10 categories of parameters (e.g., the category "longitudinal control" includes the errors in speed, time headway and time to collision). The objective of the experiment presented here was to determine the validity and sensitivity of the I-TSA tool for this evaluation process. A homogeneous cohort of 41 young, healthy males (25 to 40 years old) drove for approximately 1 hour in a static simulator environment. The scenario on a two-lane motorway consisted of 4 counterbalanced drives with easy and difficult road shapes and traffic conditions. The trial included several interaction tasks with IVIS and ADAS differing in their stage of integration and adaptivity. The successful induction of high workload levels could be both detected by objective (such as speed compensation) and subjective measures (questionnaire). Highly significant differences in the safety levels were found between the easy and the difficult drives (demonstrating the suitability of the procedure) as well as between the sections with default and integrated behavior of the information systems (supporting its sensitivity). Preliminary results support the possibility of discriminating between visual and cognitive workload, as well as sensitivity to learning effects.