Seite - 127 - in Maximum Tire-Road Friction Coefficient Estimation

7. Summary According to the relevant literature, advanced driver assistance systems (ADAS) pro- vide a high potential to increase road safety and are therefore the focus in the European white paper on road safety for 2020. The warning and activation strategies of ADAS could be significantly improved by the knowledge of two basic parameters. First, knowl- edge about the driver behaviour in the pre-collision phase can be used to optimally plan further ADAS interventions. This includes the prediction of whether or not the driver reacts to an ADAS warning and, in case the driver intervenes, it depends on whether it is a braking or steering input, for example. Secondly, the conditions of the road surface and the mounted tires limit the transmittable forces between tire and road and thus also the inputs to steering system, powertrain or braking system performed either by the driver and/or an ADAS. Currently, ADAS are designed to meet requirements for dry road, in order to avoid false interventions and reduce missed interventions. Enhancing theactivationstrategyofanADASbyanestimateof thecurrentroadandtireconditions increasesthepotential toavoidanaccidentorreducethe injuryseveritywhenanaccident is inevitable on low friction surfaces. Studies have shown that an additional potential to avoid an accident or reduce the injury severity when an accident is inevitable can be reached by adapting the ADAS intervention strategy to the current road condition. Thispotential increaseswithdecreasingmaximumcoefficientof frictionbetweentireand road, which is referred to as friction potential. The present thesis dealt with methods that identify thecurrent frictionpotential ina fast, reliableandsufficientlyaccurateway. An algorithm was developed to estimate the friction potential between tire and road, which is the primary indicator for the road and tire conditions. By using an automated emergency braking system (AEB) as an application case, requirements for the estimate were defined. One requirement is the accuracy of the friction potential estimate, which limits the ability to adapt the activation strategy. In addition, vehicle states that are most sensitive to a change of the road and tire condition were identified, as they promise to contribute to an observer of a friction potential estimate. The algorithm developed based on these prior investigations, which is mainly based on signals from standard