Modeling and dynamic analysis of a 6 x 6 heavy military truck by adaptive model predictive control with application to NATO lane change test course

摘要

In this paper, the computer-aided vehicle dynamic analysis of a 6x6 heavy military truck is presented and examined. For the analysis, a MATLAB/Simulink (R) platform is used to design and model a truck. The vehicle configuration taken into account for the analysis is the powertrain (engine, gear box, transfer gear, differential), suspension, steering system and tire model according to the Pacekja 89' formulation. In addition, the effect of the rolling resistance and drag is considered, in order to represent the vehicle behavior as real as possible. The longitudinal dynamic and lateral dynamic are formulated. First, the longitudinal dynamic model is established by means of implementation of the weight transfer function. The vehicles are considered as rigid bodies with 1 degree of freedom. Second, the vehicular planar model with three wheels, well known as bicycle model, is applied following the North Atlantic Treaty Organization double line change maneuver test reaching 3 degree of freedom. The driver behavior is represented by using an adaptive model predictive control varying the longitudinal velocity. The forces for braking, inertia of the rotating components, the energy lost in the powertrain, and the effect of dive squat and rollover. The numerical simulation results are shown and compared with a full-vehicle model formed by using Mechanical Simulation Corporation's truckSIM (R). There were chosen simulation scenarios applied to the model to observe the effects of different parameters concerning the dynamic behavior, and also prepared in truckSIM (R) environment. The main contributions of this article are the development of the vehicular model, through the use of block diagrams in a reliable and relatively simple programming code such as MATLAB/Simulink (R), with innovative tools used in the control of autonomous vehicle driving and the flexibility to adapt said model to different environmental conditions and different vehicle parameters.