Article Preview
Top1. Introduction
Active safety systems and passive safety systems are changing the vehicular technology landscape. Unfortunately, the rate of fatalities has remained near to 40,000/year for the past 15 years (Caveney, 2010) even though vehicular active safety systems and passive safety systems are very popular. In the research report, 80% of driver related accidents are due to lack of attention by the driver. More than 10 million people are injured in vehicular accidents (Cui, Liu, Li, & Jia, 2010). Unfortunately, this includes two to three million severely injured and 0.4 million fatalities in vehicular accidents (Mano, Stein, & Shashua, 2004). The Forward Collision Warning System (FCWS) has become a popular research topic in recent years (Kusano & Gabler, 2012; Nakaoka, Raksincharoensak, & Nagai, 2008; Takada, Hiraoka, & Kawakami, 2011). FCWS mostly uses radar and image processing to detect an imminent collision. The FCWS can not only detect immediate forward collision danger but also provide a timely warning for the most common causes of accidents in modern traffic.
There are many research papers focused on FCWS. Dagan et al. propose a vision-based FCWS for highway safety (Mano, Stein, & Shashua, 2004). This paper computes the time-to-contact and possible collision course directly from the size and position of the vehicle in the image. However, the accuracy of the information depends on the weather. Cui et al. propose a new vehicle detection method: appearance-based hypothesis generation with template tracking-based hypothesis verification, which can remove false positive detections, and automatic image matting for detection refinement for FCWS (Cui, Liu, Li, & Jia, 2010). However, the accuracy of camera calibration for computing headway-distance and time-to-collision also depends on the weather. Nakaoka et al. designed an FCWS based on the road friction coefficient and driver characteristics (Nakaoka, Raksincharoensak, & Nagai, 2008). The proposed FCWS cannot provide warning about emergency braking events by the vehicle in front. Takada et al. propose deceleration for collision avoidance as an index to evaluate a collision risk against forward obstacles (Takada, Hiraoka, & Kawakami, 2011). However, the proposed FCWS also cannot provide warning about emergency braking events by the vehicle in front.
In this paper, we propose a Dedicated Short Range Communications (DSRC) system combined with radar detection for the FCWS mechanism. The DSRC devices are expected to be installed in vehicles to enhance road safety. Emergency braking warning for safety messaging applications provided by DSRC is the largest step in the vehicles (Ma, Zhang, Yin, & Trivedi, 2012; Jiang, Taliwal, Meier, & Holfelder, 2006; Chen, Jin, & Regan, 2010; Biswas, Tatchikou, & Dion, 2006). The steps of the proposed mechanism are as follows. First, the proposed mechanism can actively probe the emergency brakes of the vehicle in front, relying on the relative speed and distance, and calculate the braking time between the lead vehicle and the host vehicle by radar detection. Second, the vehicle will broadcast the warning event and Global Positioning System (GPS) information when the braking time is very small. Third, GPS information on surrounding vehicles will be based on this message, allowing the braking time of the host vehicle to be calculated using this GPS information. Hence, the proposed mechanism can reduce the response delay time of the driver.
The remainder of this paper is organized as follows: Section 2 briefly reviews the background and related works. Section 3 presents the proposed DSRC, combining radar detection with the FCWS mechanism. Section 4 presents the results of experiments in the real world. Section 5 provides the concluding remarks.