Friction parameter based friction potential estimation

• Applicant: Professor Dr.-Ing. Steffen Müller
• Subject Area: Traffic and Transport Systems, Intelligent and Automated Traffic
• Term: from 2019 to 2022
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 422372111

The maximum coefficient of friction between tire and road, or friction potential, is an essential property for vehicle handling. It determines how much a vehicle can accelerate or brake and how fast it can drive in curves. The friction potential depends on a variety of friction parameters. In particular, the road condition, which describes whether the road surface is dry, wet or icy, is an important influencing factor. Other import friction parameters are e.g. the roughness characteristics of the road surface or the tire characteristics. Today, the friction potential is estimated by the driver through visual inspection based on the driver’s experience. However, sometimes there are situations, where visual inspection leads to misinterpretations, e.g. for black ice in curves or during the night. Moreover, the knowledge of the friction potential is very important for many Advanced Driver Assistance Systems, like cruise control, distance control or emergency braking, or Autonomous Driving. Although in the past numerous investigations about the determination of the friction potential have been conducted, this property can still not be identified without expensive additional sensors or relatively high wheel slip. Therefore, the question is if it is possible to determine the friction potential only by means of present series car sensors and other disposable information. Since this question has not been answered yet it is the focus of this project proposal. Within the project framework many test drives with brake measurements are performed in order to find the experimental friction potential. The sampled data is stored in a data base and each measured friction potential is correlated to different friction parameters. This data is the basis for the development of the friction potential estimation algorithm. For this, it is analyzed which friction parameter in which quality must be known to ensure an accurate and reliable determination of the lower and upper bound of the friction potential. It is also investigated which information is necessary to identify the friction parameters and how this information can be combined to increase the reliability of the friction parameter determination. For this purpose, two concepts using logistic regression and artificial neural networks are developed in order to estimate the road condition at the car’s position based on series car sensors and disposable weather data. The resulting algorithm for estimating the friction potential is implemented in a numerical demonstrator and it is tested with respect to estimation quality und robustness.

• DFG Programme: Research Grants