| Semi-active suspension control modulates otherwise passive damper forces so as to improve suspension performance. Several clipped optimal control strategies are examined and compared to a skyhook benchmark for ranges of vehicle speeds, specific points on the vehicle body targeted for improvement, and track types. Like skyhook, these control strategies first create desired control forces using linear techniques. The linear models either incorporate the primary performance metric, absorbed power, or further information regarding the vehicle model into the generation of desired control forces. Then these control forces are generated as best as possible with magnetorheological (MR) dampers incorporated into nonlinear vehicle models. The benchmark skyhook damping scheme is shown to be particularly robust, with only slight improvements across the entire performance envelope being available through alternate damping schemes. Significant improvements over skyhook are only available at localized design points. Adjusting feedback gains to the operating conditions---such as vehicle speed and test track---makes significant differences in the performance of all damping schemes, including the benchmark skyhook case. After this primary method of improvement, adding more feedback containing information about the motion and position of the vehicle body is the best way to improve skyhook damping. |
