Seite - 130 - in Maximum Tire-Road Friction Coefficient Estimation

7 Summary step) and different road conditions on the left and right wheels (µ split). In addition, different strategies for re-initialising particles after convergence were discussed in terms of noisiness and estimation accuracy. The results show that for a range of driving states, especially urban applications, the estimate of the particle filter is accurate enough to be used in an AEB. In general, the accuracy of an estimate of the friction potential in- creaseswitha longitudinalaccelerationthat is closer to thephysical limits. Forexample, this is the case during acceleration and deceleration manoeuvres with high longitudinal accelerations, as well as when driving on surfaces with lower friction potential. In the case of a certain value of a longitudinal acceleration, lower friction potentials are easier to be identified than dry conditions. The present thesis has introduced a novel method for the identification of the current friction potential that will have valuable benefit when used to enhance the activation strategies of advanced driver assistance systems, such as AEB systems. In order to achieve the necessary reliability for automotive applications, it is recommended that this approach be combined with other methods that are not based on vehicle dynamics in a sensorfusion system. Nevertheless, it is important to recognise that the proposed approach has the potential to contribute to a sensorfusion system. Compared to identi- fication methods that are not based on vehicle dynamics (e.g. measuring precipitation type or thickness, infrasctructure-aided systems), the proposed approach makes it pos- sible to observe the tire condition in combination with the road condition, which also enables thedetectionof an influenceon the frictionpotential (e.g. tirewearandageing). The proposed identification method is also suitable for identifying the friction potential during combined longitudinal and lateral excitation, since combined tire forces are con- sidered. In principal, lateral tire forces could also be used for friction estimation with the proposed approach, although the observed model would clearly need to be adapted. Nevertheless, the proposed algorithm seems to be accurate enough to be applied in an ADAS such as the AEB for urban applications, as well as for the detection of lower friction potentials than dry road during typical driving states. 130