Stopping distance is considered to be an important factor for providing a safe roadway network to the road users. Stopping distance is defined as the distance that the vehicle travels before it comes to a complete halt by considering both, the braking distance and the brake reaction/perception distance. With the emergence of Autonomous Vehicles (AVs), new challenges regarding stopping distance are currently posed to the engineering community. Depending on their total computational power, AVs can have a quicker perception as compared to human beings which can be standardized. AVs need a well-defined process to estimate the braking distance as they significantly lack the experience of human drivers. Current AI technology utilized by AVs incorporates a constant coefficient of friction (µ) for estimating braking distance and applies various factors of safety for unfavourable driving conditions. Such practices can overestimate the braking distance or sometimes underestimate it. Either estimates are not good for a smooth travel of the autonomous vehicle. This is because the stopping distance depends on the fundamental friction relationship between tire and pavement. A research study considering input of realistic tire-pavement friction, signified by the variable coefficient of friction into AVs AI toolkits has never been performed. The absence of research with respect to realistic coefficient of friction for AVs can be attributed to the fact that tire-pavement contact is a complex interaction to investigate and model.
The goal of this MSc graduation thesis is to develop a methodology for investigating the variable coefficient of friction between tire-tread rubber blocks and aggregate blocks utilizing a Tribometer. A Finite Element Model (FEM) of the interaction between tire-tread rubber blocks and aggregate blocks is developed to upscale the empirical findings from Tribometer tests to a large range of variables. The data obtained from the scientific methodologies is suggested to be incorporated into intelligent stopping distance algorithms in AVs AI toolkit to improve road safety. Such research involving Tribometer to study the coefficient of friction between tire-tread rubber and aggregates has never being performed. The study also investigated the hysteretic friction in tire-tread rubber which has never been performed previously. Key aspects of the thesis project included studying the effect of varying rubber temperature and frequency on the sliding friction between tire-tread rubber blocks and aggregate blocks. The research studied the varying effect of rubber temperature and frequency on the coefficient of friction for two surface conditions, dry/wet and three types of aggregates: Bestone, Scottish granite, cement concrete. The FEM results offered a preliminary insight on the investigated variable coefficient of friction. The scientific findings from the model and empirical results show a dependence of thermomechanical properties of the rubber material and surface conditions on the coefficient of friction between tire-tread rubber blocks and aggregate blocks. The findings indicate that the coefficient of friction decreases with increase in rubber temperature, interaction frequency and lubrication at the contact.