Road Profile Estimation and Preview Control For Low-Bandwidth Active Suspension Systems

Abstract

Active suspensions in vehicles can apply a vertical force between the wheel and the vehicle body in order to reduce vehicle body motion over road obstacles, enhancing ride comfort. In this study, new methods are developed to generate a preview road height profile using vehicle sensors and to control the active suspension using this information. To control the actuators, two concepts are developed: a feedforward disturbance compensation, calculated from the preview signal, combined with a feedback loop without preview and a preview model predictive control approach. After defining a desired road height profile as input for the designed control algorithms, a sensor-fixed and an inertial coordinate system to compile the road height profile from sensor measurements are proposed. Methods to transform the sensor data from one time step to another, to accumulate redundant measurements, and a filtering technique to shape the road signal are developed. To test the algorithms, a simulation from the road signal generation to the suspension control is designed, and additionally, an observer is derived to determine the road height profile from vehicle motion in order to verify the preview signal. Simulation results and measurements of vehicle implementation are presented.