Semi-Active Control Laws - Optimal and Sub-Optimal

Abstract

This paper attempts to clarify the question of what the optimal semi-active suspension is that minimizes a deterministic quadratic performance index. The optimal control law is a time-varying solution that involves three related Riccati equations. The constant Riccati solution (the so-called “clipped optimal” solution) is not optimal, although its performance is generally quite close to that of the time-varying solution. As the time-varying solution cannot be practically implemented, several constant gain sub-optimal solutions are investigated. A new semi-active algorithm, called the “steepest gradient” algorithm, is developed and its performance is shown to be superior to that of the “clipped optimal” solution.