State and parameter estimator design for control of vehicle suspension system

Abstract

Modern vehicle stability and navigational systems are mostly designed using inaccurate bicycle models to approximate the full-car models. This results in incomplete models with various unknown parameters and states being neglected in the controller and navigation system design processes. Earlier estimation algorithms using the bicycle models are simpler but have many undefined parameters and states that are crucial for proper stability control. For existing vehicle navigation systems, direct line of sight for satellite access is required but is limited in modern cities with many high-rise buildings and therefore, an inertial navigation system utilizing accurate estimation of these parameters is needed. The aim of this research is to estimate the parameters and states of the vehicle more accurately using a multivariable and complex full-car model. This will enhance the stability of the vehicle and can provide a more consistent navigation. The proposed method uses the kinematics estimation model formulated using special orthogonal SO3 group to design estimators for vehicles velocity, attitude and suspension states. These estimators are used to modify the existing antilock braking system (ABS) scheme by incorporating the dynamic velocity estimation to reduce the stopping distance. Meanwhile the semi-active suspension system includes suspension velocity and displacement states to reduce the suspension displacements and velocities. They are also used in the direct yaw control (DYC) scheme to include mass and attitude changes to reduce the lateral velocity and slips. Meanwhile in the navigation system, the 3-dimensional attitude effects can improve the position accuracy. With these approaches, the stopping distance in the ABS has been reduced by one meter and the vehicle states required for inertial navigation are more accurately estimated. The results for high speed lane change test indicate that the vehicle is 34% more stable and 16% better ride comfort on rough terrains due to the proposed DYC and the active suspension system control. The methods proposed can be utilized in future autonomous car design. This research is therefore an important contribution in shaping the future of vehicle driving, comfort and stability.