Lap time simulation for a Formula Student car

Abstract

Formula Student is a competition where universities all over the world compete in who can design and build the best Formula-style car. The car is judged based on design and performance in both static and dynamic events. Simulating the car’s performance prior to these events is essential in order to build a car that is competitive. The purpose of this thesis is to build a lap time simulator that satisfies the needs of Chalmers Formula Student in ways that already existing commercial simulators can not. The result of this paper is a lap time simulator that has an easy to use user interface that enables a wide option of different simulations. The user can simulate all the dynamic events at the same time and also analyze, in multiple ways, how changes on the car affects the results in these events in both lap time and subsequent points scored. This report describes the theories behind the implementation of the simulator as well as the methods used to attain the result. The initial version of the lap time simulator uses a quasi steady state pointmass model, as it is a simple, quick to simulate, and simple to develop while still keeping good accuracy to real world data. A bicycle quasi steady state model was also developed to increase accuracy at the expense of simulation time. This model includes front and rear axles and therefore also longitudinal load transfer. As the simulation time for the bicycle model is about 10 times that of the point-mass, both are options in the user interface. As the Formula Student competitions include several dynamic events to test the car’s on-track performance, it is also possible in the developed lap time simulator to simulate these events, either one by one, or multiple at once. Simulations featuring parameter sweeps were also developed to facilitate easy quantification of changes in parameters.