Seite - 15 - in Maximum Tire-Road Friction Coefficient Estimation

1 Introduction In particular, the aim of this thesis is to answer the following research questions: •Which state variables are sensitive to the friction potential in different driving states? • Based on the identified set of sensitive variables, is it possible to develop an esti- mation algorithm for the friction potential to meet the accuracy and time delay requirements for adapting the intervention strategies of an AEB (as an exemplary ADAS)? In order to answer these research questions, the present thesis is structured as follows: Chapter 1 provides an overview on the significance of the friction potential for traf- fic safety and discusses the adaptation of ADAS and VDC activation strategies to the current road conditions. Finally, AEB systems are selected as a test case, and the re- quirements concerning the tolerable deviation of an estimate of the friction potential to be used within this system are presented. Chapter 2 begins with a definition of the friction potential and then discusses the relevant mechanisms, including rubber physics and the tire/road contact. An overview on the state of the art on friction estimation is also provided. Chapter 3 shows the derivation of the necessary complexity of the vehicle and tire model to describe the tire/road contact. These models are then described in detail, as they are used for the sensitivity analysis in Chapter 4 and parts of the models are also used within the observer in Chapter 5. Chapter 4 presents the sensitivity analysis used to answer the first research ques- tion. After a description of the method used, the results are presented which answer the question about which variables are affected the most by the friction potential. This part of the thesis contributes to the systematic understanding of the influence of the friction potential on a vehicle’s dynamics states. Thus, variables are identified that are potentially suitable for use in identifying the friction potential in an observer model. Chapter 5 describes a non-linear observer which is used to estimate the friction po- tential based on the sensor signals of an ESC-equipped vehicle. After a short description of particle filtering, the relevant observer model is presented, which was developed based 15