One of the priorities of today’s research and development in the automotive industry is to enhance the safety of any vehicle; which is achieved by implementing anti-lock braking systems. Developments to enhance vehicle safety are through extensive research, simulations, and testing. This research focuses on the development and performances of a model of a tractor-semitrailer in straight-ahead braking with all brakes functioning correctly, as well as the behavior of heavy truck braking with selected brakes disabled.
The experimental field tests were performed at the Transportation Research Center (TRC) in East Liberty, Ohio. The test involved a 2006 International 9400i 6x4 conventional tractor pulling a 2000 Trailmobile 2-axle 48 foot flatbed semitrailer. The tests included lightly loaded (LLCW), half payload, and full payload, a total of three loading configurations. The tractor-semitrailer was fully modeled through TruckSim software; the vehicle’s ABS model, braking behaviors, and maneuvers were tuned to accurately reproduce the behavior of the experimental tractor-semitrailer.
The research started with building the tractor-semitrailer model in TruckSim software. Measurements were taken from the International 9400i 6x4 conventional tractor and 2000 Trailmobile semitrailer at the time of the experiment. Some dimensions used in this research required careful estimation. The estimation measurements were scaled with reference to a similar Peterbilt 6x4 tractor. In the three loading configurations, payloads were being applied in the half payload and full payload conditions. In the simulation vehicle model, payloads were applied by estimating the physical dimensions and placements on the experimental tractor-semitrailer. During the experiments, test data was collected on two computers located on the tractor-semitrailer. The test data were used to analyze the simulation models in TruckSim vehicle dynamics software.
The braking system in the simulation was developed through repeated simulations and analysis. Simulated brake control (primary treadle) was input using an actual experimentally measured brake control signal from the tractor-semitrailer at each loading configuration. Brake torque used in the simulation was based on a previously developed brake torque model that depends on brake chamber pressures and vehicle speeds. The brake torque was then adjusted during the tuning process. The ABS model was tuned by the observations and comparisons of wheel slips, stopping distances, and other braking parameters between the simulations and experiments. Disabling brakes was achieved by inputting zero brake torque in the selected brakes axle.
The simulated models were found to accurately model the braking behavior of the field test vehicle at lightly loaded, half payload and for some conditions at full payload. Comparisons between the simulated model and experimental data were made by comparing several parameters, including vehicle longitudinal acceleration, vehicle wheel speeds, brake chamber pressure, longitudinal wheel slip, yaw rate, hitch articulation, steering wheel angle and torque. The simulated model provides excellent results in the lightly loaded and half payload configurations by capturing similar vehicle braking behavior and accurately estimating the stopping distances. In the full payload configuration, the simulated model still provides a moderately accurate simulation. Yet simulations at the fully loaded condition estimate stopping distances less accurately then the other two loading conditions in most simulation runs, but closely correlate the braking behaviors with the experimental tractor-semitrailer. The inaccuracy is correlated to the full payload forces applied to the tractor and semitrailer. From the trend of the simulation results, the simulation result goes from accurate to less accurate as the payload get heavier; which indicates that more payload carried by the tractor-semitrailer, more uncertainties are involved, causing the simulated model’s accuracy to be reduced. Additional suggested studies are discussed in the final chapter of this report.