Modeling and control design for half electric vehicle with wheel BLDC actuator and Pacejka's tire

We are considering the problem of controlling the longitudinal motion of half-vehicle electric vehicle (EV) or hybrid electric vehicle (HEV). A substantial part of the problem lies in the modeling of the controlled system dynamics, including those of: (i) the half-vehicle chassis motion; (ii) the actuator constituted of a brushless direct-current (BLDC), as an in-wheel motor, along with the associated voltage inverter; (iii) the tire. The dynamics of the latter are presently captured by Pacejka's model which is known to be more most accurate as it accounts for the various phenomena generated at the contact tire/road. The whole system model also takes into account other nonlinear effects, e.g., including the road load, the rolling resistance and the aerodynamic effects. Compared to those used in existing comparable works, the system model developed in this study is much more accurate. But it is also more complex, because it is of higher dimension and stronger nonlinearity, increasing substantially the complexity of the control design problem. The control objectives include tight regulation of the vehicle's speed in the various driving modes and safety of the on-board passengers. These objectives are technically reformulated in terms of reference trajectories, to be matched by the vehicle speeds despite unpredictable variations in aerodynamic forces and road parameters. This challenging control problem is dealt by developing a cascade-structure adaptive controller. The cascade structure is resorted to make the compensation for speed dynamics and disturbances decoupled from those of torque dynamics. The adaptive nature is realized by using parameter estimators, providing the controller with the capacity of compensating the effect of parameter uncertainty. The various controller components are designed making use of several different theoretical design tools including the backstepping control method and Lyapunov stability method. In addition to theoretical analysis, the performances of the developed controller are highlighted by simulation results in the presence of varying aerodynamic forces, road slope and various road surfaces. Comparison with simpler controller is also performed.