Multi-Objective Optimization of a Path-following MPC for Vehicle Guidance: A Bayesian Optimization Approach

This paper tackles the multi-objective optimization of the cost functional of\na path-following model predictive control for vehicle longitudinal and lateral\ncontrol. While the inherent optimal character of the model predictive control\nand the direct consideration of constraints gives a very powerful tool for many\napplications, is the determination of an appropriate cost functional a\nnon-trivial task. This results on the one hand from the number of degrees of\nfreedom or the multitude of adjustable parameters and on the other hand from\nthe coupling of these. To overcome this situation a Bayesian optimization\nprocedure is present, which gives the possibility to determine optimal cost\nfunctional parameters for a given desire. Moreover, a Pareto-front for a whole\nset of possible configurations can be computed.\n