Examining Speech-Based Auditory Alerts for Intersection Collision Warning Systems using a Driving Simulator

Author(s):  
West M. O’Brien ◽  
Xingwei Wu ◽  
Linda Ng Boyle

Collision warning systems alert drivers of potential safety hazards. Forward collision warning (FCW) systems have been widely implemented and studied. However, intersection collision warning systems (ICWS), such as intersection movement assist (IMA), are more complex. Additional studies are needed to identify the best alert for directing the driver toward the hazard. A driving simulator study with 48 participants was conducted to examine three speech-based auditory alerts (general, directional, and command) in a simulated red light running (RLR) collision scenario. The command alert that informed the drivers to brake was the most effective in reducing the number of collisions. The post-drive questionnaire showed that drivers also rated the brake alert to be best in terms of interpretation (based on the Kruskal Wallis test). This study provides insight into the performance of different types of speech-based alerts for an intersection collision warning system and can provide guidance for future studies.

2014 ◽  
Vol 2014 ◽  
pp. 1-9 ◽  
Author(s):  
Ki-Yeong Park ◽  
Sun-Young Hwang

We propose a range estimation method for vision-based forward collision warning systems with a monocular camera. To solve the problem of variation of camera pitch angle due to vehicle motion and road inclination, the proposed method estimates virtual horizon from size and position of vehicles in captured image at run-time. The proposed method provides robust results even when road inclination varies continuously on hilly roads or lane markings are not seen on crowded roads. For experiments, a vision-based forward collision warning system has been implemented and the proposed method is evaluated with video clips recorded in highway and urban traffic environments. Virtual horizons estimated by the proposed method are compared with horizons manually identified, and estimated ranges are compared with measured ranges. Experimental results confirm that the proposed method provides robust results both in highway and in urban traffic environments.


Author(s):  
Xingwei Wu ◽  
Linda Ng Boyle ◽  
Dawn Marshall

Forward collision warning (FCW) systems help prevent rear-end collisions by identifying and alerting drivers of threats ahead. Understanding drivers’ avoidance strategies i.e. the tendency to brake or steer is important for the design and effectiveness of these systems. A driving simulator study was performed across five US locations to examine three driver avoidance maneuvers: braking only, steering only and combined braking and steering. A log-linear analysis was used to investigate the likelihood of an avoidance maneuver given the driver characteristics (age, gender) and study location. Findings showed that drivers aged 40 years and older were more likely to use a combined braking and steering maneuver to avoid a rearend collision. Drivers from two coastal urban areas (Washington, D.C. and Seattle, WA) were less likely to choose braking only in response to FCW alerts. Younger drivers and drivers that live in more rural areas (Clemson, SC and Iowa City, IA) were more likely to select braking only to avoid a crash, which could be due to their experience in less congested traffic environment. The findings of this study provide some insights on the factors associated with various avoidance strategies among drivers. This understanding can help guide the design of future in-vehicle collision warning systems.


Sign in / Sign up

Export Citation Format

Share Document