Comparison of two methods for the design of active suspension systems

The design of controllers for active suspension systems can be formulated as an optimal control problem. The main objective of the controller is to isolate parts of the system from vibrations in other parts. Additional constraints are limited suspension travel, limited actuator force and good track-holding capability. The objective and the constraints can be used in the formulation of an optimality criterion. Two criteria are examined, with the corresponding LQ and Hoo control design methods. The methods are compared with respect to their ability to generate a controller that achieves the best performance. For the LQ method three controllers with different structures were generated. To illustrate the use of the design methods, controllers have been designed for a simple but typical suspension system. For this system the controllers based on a quadratic norm perform comparably but the controllers based on output feedback are less robust. The Hoo controller did not perform well for this problem when the same weighting functions were used to generate the controller as for the LQ method. Use of the Hoo design method therefore requires careful tuning of the weighting functions. Use of the standard functions is inappropriate.