Hybrid Optimization Problems in Automotive Applications

Abstract The design of a control for an automotive engine in the cut-off region is considered. The system is modeled as a hybrid control system, and a minimum time problem is formulated. Two relaxed problems obtained by reducing the hybrid model to the continuous and discrete time domains, respectively, are first solved, providing lower bounds on the attainable cost. Then, sub-optimal solutions to the original problem are devised by mapping the solutions of the relaxed problems back to the hybrid domain. The performance of the hybrid laws thus obtained are discussed.