This study investigated three issues relevant to Advanced Traveler Information System (ATIS) design: (1) the influence of an ATIS on driver performance in reduced visibility conditions, (2) the influence of an ATIS on drivers' reactions to unexpected roadway events, and (3) the interaction of an ATIS with a Collision Avoidance System (CAS). The study was conducted in two phases, combining a very efficient confounded experimental design used in Phase I with a traditional orthogonal design used in Phase II. Both experiments were conducted in a high-fidelity driving simulator. A total of 20 drivers was tested: 8 in Phase I and 12 in Phase II. In both phases, subjects completed several driving scenarios, during which they received roadway-relevant information via an in-vehicle ATIS and variable message signs (VMS) posted on the roadway. Subjects also experienced several unexpected roadway events, some of which triggered a CAS alert. Results showed that mean speed was lower in the ATIS condition than in the control condition, while out-of-vehicle VMS messages did not alter speed. Contrary to our expectations, effects of ATIS and VMS messages upon driving performance did not depend upon visibility conditions. Drivers in clear and fog visibility performed consistently. The results also showed that an ATIS message interfered with the driver's ability to react to a pedestrian road incursion. However, an ATIS message did not interfere with a CAS warning. The present results need to be expanded before design guidelines, especially for integration of in-vehicle systems, can be written with great confidence. However, these results do indicate that the simulator can provide a reasonable test bed for future evaluation of system interactions. (Author/publisher)
Samenvatting