Search Results

Elastokinematic properties of the suspension system are one of major factors that improve ride and handling performances. Their properties have been quasi-statically determined using the in-door measurement device, such as the SPMD (Suspension Parameter Measurement Device). The elastokinematic properties in driving conditions, called dynamic compliance characteristics, are defined as the superposition of quasi-static properties, called static compliance characteristics. However, this superposition method has difficulties in predicting the suspension non-linearities under high lateral acceleration and transient behaviors. In this paper, dynamic compliance characteristics are directly determined through on-board measurements and transformation matrix. Their applications in validating ADAMS full vehicle model are discussed. Some experimental and simulation results about their unique properties and relationships with SPMD are presented.

A determination of the vehicle states and tire forces is critical to the stability of vehicle dynamic behavior and to designing automotive control systems. Researchers have studied estimation methods for the vehicle state vectors and tire forces. However, the accuracy of the estimation methods is closely related to the employed model. In this paper, tire lag dynamics is introduced in the model. Also application of estimation methods in order to improve the model accuracy is presented. The model is developed by using the global searching algorithm, a Genetic Algorithm, so that the model can be used in the nonlinear range. The extended Kalman filter and sliding mode observer theory are applied to estimate the vehicle state vectors and tire forces. The obtained results are compared with measurements and the outputs from the ADAMS full vehicle model. [15]

In the early stage of vehicle development process, it is customary to establish a set of goals for each kinematic and compliance (K&C) characteristic and try to find out design variables such as the location of hard points and bushing stiffness which can achieve these goals. However, since it is very difficult to find out adequate set of design variables which satisfy all the goals, many engineers should rely on their own experiences and intuitions, or repeat trial and error to design a new suspension and improve old one. In this research, we develop a suspension design process by which suspension K&C characteristic targets can be achieved systemically and automatically. For this purpose, design optimization schemes such as design of experiments (DoE) and gradient-based local optimization algorithm are adopted.