Search Results

Staged motorcycle-to-car and motorcycle-to-barrier collisions were conducted with seventeen early 1990's models Kawasaki 1000 motorcycles. The impact speeds into the barrier and cars were varied between 10 and 49 MPH. The purpose was to observe the change in motorcycle wheelbase, and characterize motorcycle-to-car and motorcycle-to-barrier crush profiles. These crash tests will expand the existing motorcycle crash test database. The vehicles were instrumented with tri-axial accelerometers to facilitate the analysis of forces, speed change, and stiffness. Some of the crash tests were recorded by high-speed video cameras. This paper characterizes the data collection system, summarizes the data collected, and lists the parameters that characterize the collision. Crush data and vehicle rest positions were recorded by typical reconstruction methods.

Lateral load balance may be important in some rollover incidents and may provide a measure that could relate to the pavement damage from unbalanced loading. Individual wheel loading was measured on 92 heavy trucks operating in western Oregon during 1995. This paper presents the analysis of those results in the form of distributions for various categories of wheel loads and for several unbalance measures. Unbalance was found to have approximately normal statistical distribution with the center of gravity off-set up to 4 inches from the vehicle center. For axle sets the center of gravity off-set was up to 7.5 inches.

The data from a set of 45 reference cases, involving the collision of two generic automobile models, is presented as a basis for comparing collision algorithms used in motor vehicle accident investigation. The set was generated by the SMAC algorithm.

A controlled experiment involving road marks was conducted to compare various photogrammetry practices currently in use in the accident investigation community. The experimental controls and results are discussed for three variations of one 2-D scheme and for six 3-D photogrammetric schemes applied to a similar set of road marks and points. For measurements related to the most frequent issues in traffic accidents, all of the methods are capable of providing usable data. The experimental photographs and corresponding data represent a reference set for developing skills and for comparison with other photogrammetry schemes.

A photogrammetric transformation is presented, which maps photograph image points to the location of the same points on a nominally plane real surface. The transformation uses 4 base points on the real surface that have known locations and is implemented by an interactive program entitled TRANS4; which runs on MS/PC DOS microcomputer systems. The program addresses three of the five major tasks in the process of extracting geometric data from photographs of a traffic accident site and makes the method a realistic option for investigators. The two other tasks are also discussed. The character and relative magnitude of several errors associated with the transformation are discussed, along with techniques for managing the errors.