Validation of a two-wheel vehicle model using Genetic Algorithm

Abstract

Vehicle dynamics simulation is widely used in the automotive industry. The usability of each model depends on how well it can replicate the behaviour of the real vehicle. Each simulation model must go through a thorough investigation process- model validation- before use. Accordingly, the research introduces a comprehensive methodological framework for vehicle dynamics model validation using a sophisticated vehicle dynamics measurement system and a machine learning algorithm. The validation process of a two-wheel vehicle model (bicycle model) is presented where the authors’ previously proposed work is exploited as a base and further developed, i.e. by using a Genetic Algorithm (GA) the vehicle parameters (yaw inertia, axle cornering stiffness) are estimated in an automatized fashion.