Desenvolvimento de novo modelo veicular quase-estático e novo algoritmo de otimização para a predição do tempo mínimo de volta

Abstract

In this work, a new method for quasi-static modeling of competition vehicles was developed. The method uses only information that is readily available for amateur automotive competition categories. An optimization problem was formulated to determine the optimal trajectory and a new optimization method was developed to perform the trajectory optimization. This new method (Double Gradient Method) makes it possible to predict racing lines described by cubic splines (problems solved in most cases by stochastic methods) in a similar time as deterministic methods. Solving the minimum lap time problem using a graphical optimization method proved to be feasible and applicable. The double gradient method was evaluated against four other optimization methods capable of graphically optimizing the trajectory and was the only one that could run the simulation in a timely manner, allowing the analysis to be performed during sporting events.