This paper presents an efficient method to design and assess optimal feedforward control for active steering configurations. Impacts of additional steering angles at the axles and the related control strategies can be evaluated for specifiable target behaviour. This process eases complex engineering decisions in the early development period. The high-dimensional trajectory optimisation problem is basically solved by formulating a quadratic program. This convex description is extended to the nonlinear vehicle motions by iterative quadratic programming. The essential parts of this method are solving an optimisation problem by iterative operating point linearisation and finding solutions for arbitrary long manoeuvres by fragmentation. To describe lateral vehicle behaviour, an advanced, nonlinear bicycle model with actuation submodels is identified through measurements. The overall objective assessment method is based on a chosen central reference model fed by trajectories of open-loop manoeuvres. Investigation of steady-state and dynamic responses is performed through compact objective criteria leading to assessment webs. Differences in regard to reference tracking and robustness are evaluated for several control strategies.
|Number of pages||20|
|Journal||Vehicle System Dynamics|
|Publication status||Published - 4 Apr 2017|
- Lateral vehicle dynamics, active steering configurations, feedforward control, trajectory optimisation, quadratic programming, objective assessment, DYNAMICS, OPTIMIZATION