This report describes the test of a method proposed for assessing safety effects of modern in-vehicle information systems. In the experiment, twelve subjects followed a prespecified, urban route using a map-based route guidance system. The entire route contained three instances of four common driving situations: (1) turning right; (2) negotiating general rule intersections; (3) negotiating yield intersections; and (4) driving straight ahead at major inner-city arteries. Safety was assessed by subjective ratings and objective measurements. The results showed that: (a) the subjective ratings were more sensitive to the effects of the route guidance system than the objective measures; and (b) that non-safe events occurred over twice as often with the guidance system than in a control condition. Safety reductions were found only with respect to braking and course keeping (both control tasks) and not in the interaction with other traffic participants (manoeuvre tasks). (A)
Abstract