Dynamic curve warning systems (DCWS) and automatic truck rollover warning systems (ATRWS) are installed to mitigate speeds at locations where geometric features make it difficult for drivers to maintain a safe speed (Tribbett et al. 2000). These warning systems belong to a family of systems commonly referred to as dynamic speed feedback signs, radar speed signs, radar message signs, electronic speed signs, and dynamic speed display signs. However, ATRWS and DCWS are specifically installed on curves and other locations with higher risks of rollover crashes. Trucks are often involved in rollover crashes at curves that require reduced speeds. On downgrades, curves further exacerbate the incidence of truck crashes due to the loss of braking efficiency (Myers et al. 1981).
DCWS usually consist of a speed measuring device, such as loop detectors and radars, and a VMS that displays feedback to drivers who exceed a threshold speed. Other devices which may be incorporated into the system include cameras, video detection software, control/communication equipment, a power unit (eg. solar panels) and a computer software to manage the inputs and modify the sign on the VMS. Real-time information is communicated to drivers such as the recommended speed of the curve and the driver’s current operating speed. Other curve warning systems are installed to operate in a sequence. These can incorporate flashing chevron signs along with an advance warning sign in the horizontal curve (Smadi et al. 2014). The main advantage regarding DCWS is that they have a much greater effect on high-speed vehicles than static curve warning signs and the dynamic system significantly enhances the ability of vehicles to successfully negotiate through curves (Torbic et al. 2004). An example of a DCWS installed in Oregon is shown in Figure 1.
DCWS have been operational on several highways in states such as Iowa, California, Minnesota and Washington since the early 2000s. Evaluations on their safety effectiveness have generally been consistent with a significant reduction in excess speed-related crashes on curves being observed (Hallmark et al. 2015; Knapp and Robinson 2012; Tribbett et al. 2000).
ATRWS operate by measuring potential rollover metrics as vehicles approach a dangerous section. The most critical measure of the potential of rollover is the static rollover threshold which is expressed as lateral acceleration in gravitational units (g). Passenger vehicles predominantly have a threshold greater than 1 g, while light trucks, vans and SUVs have values ranging from 0.8 to 1.2 g (Winkler et al. 2000). The typical five-axle semi-trailer combination popular on U.S. roads has a rollover threshold only as high as 0.5g with an optimal high-density, low center-of-gravity load when loaded to legal gross weight (Winkler et al. 2000).
From the above, it is clear that heavy vehicles are more susceptible to rollover crashes than light vehicles caused by inadvertently operating the vehicle beyond the rollover threshold. ATRWS have been in development for some time. These use ITS technologies to provide an automatic assessment of rollover risks to approaching vehicles. A warning message is activated once the risk is identified. The message may be displayed on a VMS or by activating a flashing light which alerts drivers to the potential risk. Factors considered to assess rollover risk include vehicle type, speed, weight and height. A determination is made if an approaching vehicle has exceeding the estimated rollover threshold (Donnelly 2008). The basic detection and classification of the ATRWS include WIM detectors, piezo-electric or radar devices for speed detection, radar-based height detectors, and an overhead VMS to display both caution messages and advisory speeds.
Recent ATRWS algorithms incorporate additional vehicle parameters such as live load, non-live load, and vehicle configuration into the rollover threshold equation. This significantly improves the accuracy and effectiveness of the rollover warning system (Baker et al. 2001). Such an algorithm has been developed by The University of Michigan Transportation Research Institute (UMTRI)(McGee et al. 1992; Strickland and McGee 1997). Other systems enable the inclusion of road geometry into the algorithm for calculation of a safe speed (Bergan et al. 1997). ATRWS have been installed at hazardous locations prone to truck rollovers along with DCWS and have been shown to be effective (Baker et al. 2000; Bertini et al. 2006). A picture of an ATRWS used with a static warning sign is shown in Figure 2.