Active Suspension Control Synthesis Using Reduced Order Models That Account for the Time Delay Between Road Inputs

Abstract

The control of wheeled ground vehicle suspension systems is well suited for analysis and refinement using multi-input multi-output (MIMO) control law synthesis methods for linear systems. Usually it is necessary and desirable to develop the control algorithms using a reduced order model of the system. Since such vehicles are also characterized by correlated road inputs with time delay between the front and rear wheels, it is also desirable to consider this delay during the model reduction process. If this delay is taken into consideration, then it may be possible to develop low order control algorithms which compensate for the vehicle modes that are disturbed by the road inputs, resulting in improved overall performance. This paper describes the application of model reduction to a model of a ground vehicle for active suspension control law synthesis. The vehicle is described by a high order MIMO model of a “half-car” with four rigid-body degrees of freedom and flexible body modes to account for structural vibration, plus additional states to represent colored noise road disturbance inputs. Fourth order MIMO models suitable for control law synthesis are then determined using internal balancing, taking into consideration the time delay between the front and rear wheels, followed by subsystem elimination. The performance of the vehicle (high order model) with the resulting low order active suspension control laws is then assessed.