Study On Integrated Vehicle Dynamics Control Based On Optimal Tire Force Distribution

Due to the increasing number and complexities of electronic control systems in modern vehicles, Integrated Vehicle Dynamics Control (IVDC) has recently become a hit topic in the research field of vehicle dynamics. Based on the inherent coupling of vehicle dynamics , the control integration among active systems of steering, braking, traction and suspension is researched in this dissertation. The proposed integrated controllers are also validated by open-loop and closed-loop virtual tests through professional vehicle dynamics software -MSC CarSim.A full vehicle model with 14 degrees of freedom(DOF), including 6 DOF for body, 4 DOF for tire vertical hopping and 4 DOF for wheel rolling, is established for the design of integrated controllers. As the most important part of modeling, a static tire model based on Burckhardt approach is established. To get more practical tire model parameters that can be used for controllers, several virtual tire tests based on CarSim are run and the obtained tire longitudinal and lateral force data are further fitted using Levenberg-Marquardt least square algorithm. Then in CarSim, a more realistic vehicle is modeled, and particularly, four fully nonlinear independent suspensions, nonlinear tires and complex aerodynamic effects are all included.The two main goals of control integration, i.e. interference avoiding and optimized control effect allocating, can be fulfilled using a top-down integration strategy. With this strategy, a main-loop/servo-loop structured controller is proposed, in which tire nonlinearities can be directly taken into consideration within the servo-loop. Besides, different from the commonly-adopted linear feedback control method, nonlinear robust sliding mode control (SMC) theory is used to calculate the desired stabilizing forces/moments for the vehicle, i.e. F_ud. With SMC, no linearization is necessary but nonlinear dynamic coupling can be directly considered.Since tires play a vital role in controlling a vehicle, tire force distribution in the servo-loop, is undoubtedly the most crucial part of entire controller. In this dissertation, two kinds of optimization approach, i.e. unconstrained and constrained, are applied in the servo-loop to reasonably and optimally distribute the stabilizing forces/moments to multiple tire actuators.Firstly, an integrated controller for steering, braking and traction are designed using unconstrained optimization, and it is shown that by integration vehicle handling and stability performance can be greatly improved. Then active front and rear stabilizers are further integrated to actively control the tire vertical loads and thus to further improve vehicle handling. It is shown that by introducing stabilizers, vehicle roll stability can be significantly enhanced, while vehicle handling can also be improved to some degree.In order to further consider different tire actuator limits, the so-called control allocation (CA) approaches, including weighted pseudo inverse (WPI) and sign-preserving quadratic programming (SPQP) based CA, are applied in the tire force distribution. The test results show that WPI can be very effective in distributing optimal tire force even under low adhesion condition, but it still can not directly consider the physical limits of tire actuators. On the other hand, with SPQP based integrated controller, the vehicle can be still quite stable even under actuator failure occasions. It is because when actuator failure is detected, SPQP based controller can redistribute the stabilizing forces/moments to other actuators, which means that the robustness of integrated controller brings additional reconfigurability to the vehicle. Therefore, the global vehicle dynamics performance can be significantly enhanced by using the proposed integrated controller.