The evaluation of vehicle handling characteristics by measurement techniques is currently carried out by standardised and company-specific testing procedures. The aim of this thesis is to develop appropriate procedures to reproducibly determine the handling characteristics and facilitate the simplification of the conventional standard methods. First of all, a description of the state of the art for the objective evaluation of the vehicle dynamics of passenger cars and heavy commercial vehicles is given. The focus of this thesis is based on the mathematical calculation (identification) of characteristic vehicle parameters. For the identification, the utilized simulation model is based on an extended single track model. By using a Least Squares method the parameters of selected transfer functions of the vehicle dynamics can be identified. This approach brings a simplification of the standard analysis and allows the calculation of frequency responses and response times from freely driven manoeuvres in the range of linear vehicle dynamics. By using nonlinear optimization, the calculation of lateral force characteristics can be achieved, which are mainly influenced by tire properties. These quantities are independent of the driving speed and allow evaluations and simulations, also in the complex, nonlinear range of vehicle dynamics. Finally, the developed methods are applied to measurement data of a passenger car and a heavy truck. The measurement data of the driving manoeuvres has been recorded by using inertial measuring techniques and a steering robot: Steady state cornering, sinus steering, step-steer input and lane-change. The application of the presented methods shows that the model based approach allows a specific vehicle dynamics analysis and a simplification of the current standardised methods.