In the past few decades, the introduction of electronics in motor vehicles has marked its development. At the beginning, electronic systems were used to control the engine (electronic fuel-injection systems). From that time on, electronic components entered the domain of driving safety (e.g. the Anti-lock Braking System, Electronics Stability Control or the Adaptive Cruise Control) up to the point that completely new fields of application have emerged in the areas of driving assistance, communication and infotainment as a result of continuous improvements in semiconductor technology. This thesis is focused on the second component mentioned, the Anti-lock Braking System (ABS). Specifically, the ABS prevents the wheels from locking when the brakes are applied by detecting incipient wheel lock on one or more wheels and makes sure that both lateral and longitudinal friction are optimal by dynamically controlling the brake pressure of individual wheels. By doing so, wheels are prevented from locking up, the braking distance is minimized and the vehicle remains steerable. The Electronic Control Unit (ECU) contains, among others, the ABS functionality, which is comprised by two main parts: the high level ABS algorithm and the low level brake pressure control. The first sends a pressure request signal - determined from complex control systems based on heuristic rules - to the pressure controller, which has to be applied on the desired brake pad precisely. This work is focused on the low level control in order to make it as precise as possible and perform optimally with changing hydraulic system characteristics. By carrying out a wide analysis of response data with the current feed-forward controller structure, its system characteristics and key parameters have been identified. This has been possible thanks to a partnership between TU Delft and SKF, from which a BMW 5 series test vehicle has been acquired and modified for any kind of safety control system, such as the installation of active suspension, force sensing bearings or the hereby needed hydraulic ABS circuit modification. The main outcome of the first part of the work is the definition of a new model which, a part from considering the voltage as a new input for the pressure step estimation, improves the build up phase accuracy more than a 10% by smoothing the compressibility effect of the brake fluid. The second part of the work focuses on the design of an Adaptive Brake Pressure Controller which is based on an adaptive mapping continuously updated by the Recursive Least Squares algorithm. The results are quite promising. Indeed, this novel control system is expected to increase the accuracy of the initial controller more than a 40% while adapting to the changing-system, thus accomplishing the main objectives of this work. Furthermore, the smaller pressure steps, the main drawback of the previous feedback controller, are presumably going to be accurately reached. Last section of this chapter suggests different methodologies to determine the quality of the new designed adaptive control system which, if proved to be successful, would be a great step in the development of this important active safety control system which is the Anti-lock Braking System.

Outgoing