Vehicle Layout Optimization Using Multi-Objective Genetic Algorithms

Layout optimization of an autonomous electric vehicle is an important problem that needs to be addressed. Optimization decision space includes variables such as placement of motor and battery, type of doors, seating configuration etc. Objective functions chosen are passenger spaciousness and comfort, ingress/egress metrics, mass and cost. For this unconstrained optimization problem, a multi-objective genetic algorithm (MOGA) is used which converges on a set of optimal configurations from which the design decision makers can choose based on the product requirements. The optimization framework developed uses MOGA available on MATLAB. MATLAB is linked with Altair Hypermesh which handles the vehicle and subsystem CAD models. The developed framework was found to give reasonable results for the modeled vehicle configuration with five objectives that are evaluated as a combination of qualitative as well as quantitative estimation.