To improve your racecar lap times, Florian Maile and Vincent Rudolf of Global Formula Racing and Christoph Hahn of MathWorks demonstrate the benefits of vehicle dynamic control modeling with a focus on torque vectoring. Using Simulink^® and MATLAB^® code, the Global Formula Racing team, which placed third in the Simulink Student Challenge, explains the steps of controller design and the testing and tuning that follows. Formula Student teams have a limited amount of real-time testing. By efficiently using simulations, teams can reduce the amount of actual time needed on the track.

Torque vectoring is a vehicle-control system used for cornering and lateral movements, and also results in easier handling for the driver. The design goals for a torque vectoring controller are good response behavior, high accuracy, and linear behavior of the car during high loads. Florian and Vincent create the optimal controller by combining a slow PID controller for accuracy, and a faster, less accurate feed-forward controller.

Based on a single-track model, the Global Formula Racing team explains the steps for parameter identification, testing, and application, and then displays a software demonstration of parameter optimization with the car model. Using the CarMaker model GFR14e by IPG, the vehicle dynamic controls structure is entered into the Simulink model and, using their own MATLAB script, they optimize the parameters.

This process saves a significant amount of time because it enables you to tune your controller in a simulation environment instead of a test environment. You will be able to view the desired response of the car, and as proven by Florian and Vincent, the correlation between the model data and actual test data was very good for factors such as acceleration, wheel speeds, and yaw rate. Ultimately, simulations can reduce the amount of real-time testing, but it is important to make the most out of every test day.