A Concept for Estimation and Prediction of the Tire-Road Friction Potential for an Autonomous Racecar

This paper presents a concept to estimate and predict the friction potential between a vehicle’s tires and the road surface. An important aspect of this research project is a local high-resolution evaluation and storage of the obtained data to gain precise knowledge of the road conditions. Next to a current-state friction potential estimation algorithm, a concept is introduced which aims to predict the tire-road friction potential for a predefined horizon ahead of the vehicle. The concept focuses on an autonomous racecar, which will drive on different racetracks during the Roborace Season Alpha events. After a brief overview of state-of-the-art methods in tire-road friction potential estimation, the overall concept is presented and its application on the real-world racecar is outlined. The effects of the high-resolution friction potential information on the raceline trajectory planning algorithms are shown. Lastly, aspects of future work regarding research and implementation of the presented concept and a transfer to public road cars are discussed.