This thesis was written at the institute of computational mathematics of the University of the German Federal Armed Forces and investigates new approaches in the field of vertical vehicle dynamics using optimal control techniques. The objective is to optimize a quantitative cost function using an adapted control. The considered optimization criteria comprise ride comfort as well as safety and durability values. The first part of the thesis investigates a formulation using continuous dynamics and develops a machine learning based detection algorithm in combination with a sensitivity update rule that shows real-time performance. Therefore the regarded disturbances are assumed to appear as so called singular events, e.g. potholes and thresholds. Those can be parameterized using a small amount of parameters, what is then used to calculate the optimal control and the parameter sensitivities with respect to changes in the disturbances. In the second main part the dynamics are extended to impulsive systems, s.th. also contact losses between wheel and road surface can be included. This formulation is done using Riemann-Stieltjes-Integrals and a restitution law using generalized coeffcients. For this problem formulation explicit integration methods are presented and investigated. The existence of parameter sensitivities for the discontinuous system are developed and on top of that two gradient method based optimization routines are shown and their performance is evaluated for examples from the field of vertical vehicle dynamics.    «

This thesis was written at the institute of computational mathematics of the University of the German Federal Armed Forces and investigates new approaches in the field of vertical vehicle dynamics using optimal control techniques. The objective is to optimize a quantitative cost function using an adapted control. The considered optimization criteria comprise ride comfort as well as safety and durability values. The first part of the thesis investigates a formulation using continuous dynamics and...    »