A High Accuracy Full Vehicle Modeling Scheme with Air Suspension Systems

• Published in: 2022 International Automatic Control Conference (CACS) pp. 1 - 5
• Main Authors: Hsu, Wei-Feng, Lu, Po-Han, Lin, Yu-Chen, Wang, Chia-Hung, Yan, Zhi-Cheng
• Format: Conference Proceeding
• Language: English
• Published: IEEE 03.11.2022
• Subjects: air suspension systems; Atmospheric modeling; CarSim; full-vehicle dynamic model (FVDM); Geometry; kinematics and compliance testing; Mathematical models; Suspensions (mechanical systems); Table lookup; Tires; Wheels
• DOI: 10.1109/CACS55319.2022.9969838

The precision of modeling a full vehicle dynamics model plays an important role for research development and control design within the automotive industry. This paper aims to develop a high accuracy full vehicle modeling scheme with air suspension systems. First, CarSim is used to establish an approximate real vehicle model with air suspension systems as a target vehicle. To make the CarSim target model closer to the real vehicle, the kinematics and compliance testing (K&C testing) is performed. Next, a mathematical dynamic modelling and parameter identification strategy is proposed to construct a full-vehicle dynamic model (FVDM) with 18 degree of freedom (DOF), which including chassis K&C effect, such as the relationship between the vehicle attitude and toe angle, camber angle, etc., so that the proposed FVDM can approximate to the target vehicle from CarSim. The parameters of FVDM, such as installation ratio of suspension system, tire cornering stiffness, are derived by using the optimization scheme. The average accuracy of vehicle dynamic responses of the proposed FVDM compared to the target model is greater than 95%. This paper not only solves the effect of left-right asymmetric spring stiffness on vehicle yaw dynamics caused by suspension geometry changes, but also solves the problem of model inaccuracy caused by the difficulty of correctly calculating the chassis compliance deformation due to rubber joints of the suspension. Furthermore, a suitable full vehicle model can be used for design and simulation of control strategy and automated driving functions.