To provide additional on-track support for racing drivers, a novel real-time braking point estimation system was developed and tested to improve overall lap time performance. The Formula-Student competition was chosen as testing environment because of direct access to a real car and, hence, to all the data needed for this study, which can be fully disclosed at the same time. To estimate the optimum braking point, a simulation of the track section ahead is performed using a multidimensional approach to model not only longitudinal and lateral acceleration, but also combined acceleration scenarios. A lap identification system as well as live track analysis tools were developed in order to obtain track data that cannot be measured directly. To evaluate the overall grip conditions, a tire model, using both a wheel load and a two-track model, is implemented, which includes a dynamic calibration feature designed to adapt the system to the changing conditions of the circuit. It can be seen that the simulation provides a detailed profile of the maximum permitted velocities for every point in front of the real vehicle, and that this can be used to create a reliable prediction of the braking points along the track. Furthermore, a concept based on physiological fundamentals was developed and investigated in terms of human reaction times in order to meet the requirements of an intuitive human-machine interface. Our results demonstrate that it can be successfully used to automatically compensate for driver delays, improve the trust in the system and ensure safe operation. We conclude that our new system can reduce lap times, depending on the length and number of curves, by a significant amount of time (tens of a second).