The objectives of this study were to perform computer simulations of vehicle-curb and vehicle-berm impacts, to characterize the behavior of a wide range of vehicle types after such impacts, and to produce design and evaluation trajectory data for use by NJDOT engineers. The impact simulations performed involved a wide variety of vehicle types and several different curb and berm configurations that are typical of those in use in the state of New Jersey. Simulation results from this research, primarily in the form of vehicle bumper trajectory plots, were produced to supplement existing curb-impact vehicle trajectory databases. Vehicle trajectory data of this type is typically used to determine appropriate set-back distances for guide rails(railings) that are located near curbs. Such railings must be positioned so that vehicles impacting curbs do not overshoot the top of railings placed nearby.

Due to the wide variety of curb and berm profiles used in New Jersey and due to the even wider variety of vehicle types traveling our roadways, a large number of impact simulations were performed for this project in an attempt to cover an adequate spectrum of possible impact scenarios. Six different vehicle types- including vehicles ranging from compact cars to minivans and sport utility vehicles- were simulated impacting several different curb and berm profiles. In addition, for each vehicle and curb combination, the impact simulations were performed for several different impact angles and impact speeds. To account for possible variations in vehicle suspension characteristics, a range of vehicle suspension values were used for each vehicle simulated. After performing the impact simulations using suspension values at both ends of the chosen range of values, an envelope of possible vehicle trajectories was generated from the simulations results.

The research approaches employed in this project consisted of using numerical simulation techniques to perform vehicle impact analysis. These techniques were the HVOSM (highway vehicle object simulation model) method and the FEA (finite element analysis) method. The HVOSM system represents a vehicle as a relatively small number of discrete objects, each having lumped mass an inertial properties, and each being connected to other parts of the vehicle through links. Vehicle and tire properties for use in the HVOSM simulations were obtained from several different sources available in research literature. In the FEA method, a fundamentally different approach is used. Rather than representing the vehicle by a small number of “lumped” objects, the FEA approach is to model the vehicle as a large collection of very small pieces (or elements). Each element accounts for only a small portion of the vehicle and the properties of each element represent the properties (e.g. tire stiffness, steel stiffness, etc). of that small portion of the vehicle. These elements are then linked together into a large model, typically on the order of several thousands to tens of thousands of elements in size. Each of these methods, i.e. HVOSM and FEA, offer some advantages and disadvantages. These issues are discussed in detail in this report.