|
Abstract
|
27th International Technical Conference on the Enhanced Safety of Vehicles (ESV): Enhanced and Equitable Vehicle Safety for All: Toward the Next 50 Years
https://www-esv.nhtsa.dot.gov/Proceedings/27/27ESV-000111.pdf
In everyday driving situations, potential sources of collisions can appear from any direction around the driver. Driver assistance systems have been highly desired to assist driver's hazard awareness from all directions in order to eliminate any kinds of traffic accident fatalities. The current study addressed whether simulated spatial sounds providing directional and hazard attribute cues for potential collisions can facilitate drivers' identification of traffic hazards and reduce collision incidence in the front and rear spaces. Forty-eight participants took part in the simulator experiment. The authors used a driving simulator (Honda Driving Simulator Type-DB Model S) to present them various traffic scenes with respect to the hazardous direction and recorded their driving operations. Participants' gaze directions were also recorded with an eye tracker implemented on the simulator. To provide a directional cue of hazardous traffic participants, the authors presented spatial sounds on the directions of hazard participants, using two speakers implemented in a driver's seat. To provide an attribute cue for hazardous objects, the authors classified the traffic participants into four categories (vehicles, motorcycles, bicycles, pedestrians) and presented a corresponding imitative sound for each hazard object. Presentations of monaural sound without directional cue and signal sound without attribute cue were also used as a comparison basis. The current study observed a decrease in collision frequency and a significant reduction of onset time for pushing down the brake pedal for frontal hazard when spatialized signal sounds were presented compared with no HMI condition. A decrease in collision frequency with gazed hazards in the rear space was also observed when spatialized imitative and signal sound were presented relative to no HMI condition. The results lend to support the hypothesis that the directional cue can be effective for safer driving behaviors. On the other hand, improvements were not obtained when attribute cues were presented for both behavioral responses or the collision frequency. Significant facilitations were found in gaze responses and decelerate operations especially for rear hazards, but they did not result in a reduction of collision frequency. Although the well-known front-rear ambiguity was confirmed in stationary sound localization, the current study observed the effectiveness of directional cue in reducing the collision frequency. It is possible that movements of spatial sound sources with hazard traffic participants could improve the resolution of front-rear sound localization. The influence of front-rear ambiguity might have also been reduced by extended spatial attention from the rear to Maruyama 2 the front under the auditory directional cue towards the rear space. The attribute cue did not provide any effective improvements in the current study. However, the authors believe that in certain traffic situations where the type of hazards involved could represent more important information to the driver, the effects of attribute cue could reveal a potentially larger impact. The observations of the effective assistance of directional cue in spatial sound provide important references in terms of human factors for considering informative HMI that facilitates hazard awareness from all directions and help safer driving behaviors.
Language: en |