Active chassis components and therefore control strategies are increasingly attached to today's vehicles. Systems such as active front wheel steering (AFS), torque vectoring (TV) or active rear wheel steering (ARS) can autonomously influence the driving behaviour. This offers the possibility of improving vehicle handling and stability. However, the driving behaviour is highly dependent on the performance of the control strategy. Low control performance introduces phase lags and leads to irregular vehicle responses. High control performance is often associated with reduced control robustness and the risk of instability. In summary: in order to improve vehicle dynamics with active chassis systems, controllers must provide high levels of performance and robustness at the same time. A new control strategy was developed on the basis of the trade-off between controller performance and stability. This strategy consists of a combination of model-based identification and feed-forward control. While driving, the model-based identification estimates time-variant enhanced cornering stiffness which can be used to map a single- track model piece by piece to actual driving behaviour. The driving behaviour can then be analysed with methods of linear system theory. In addition the enhanced cornering stiffness can be used to adapt the feed-forward control which leads to steadystate accuracy.
Real-time model identification - a new approach for controlling vehicle dynamics with rear wheel steering
2011
12 Seiten, 9 Bilder, 6 Quellen
Conference paper
Storage medium
English
Automotive engineering | 2011
|