Design and Implementation of Torque Vectoring for the Forze Racing Car

Design and Implementation of Torque Vectoring for the Forze Racing Car

Stoop, A.W.

Holweg, E.G.M. (mentor)

Mechanical, Maritime and Materials Engineering

Delft Center for Systems and Control

2014-07-11

The Forze Hydrogen Racing team in Delft is developing the Forze VI; a hydrogen powered racing vehicle. To improve the vehicle dynamics for the Forze VI a torque vectoring approach is roposed. Torque vectoring provides vehicle stability control, for both vehicle accelerating and decelerating conditions by transmitting vectoring torque between the left and right wheels without deteriorating the longitudinal performance of the vehicle. In this thesis two di?erent torque vectoring control algorithms will be presented. The ?rst control algorithm is a PID and LP controller based upon a linear single-track model. The second controller is a LP controller based upon a two-track model. Both controllers are yaw rate tracking controllers. In this thesis it is assumed that a neutral steered vehicle model provides the desired yaw rate. This neutral steered yaw rate is then imposed on the Forze VI by both controllers. The dynamics of a vehicle can be very complex. The vehicle dynamics depend upon various parameters and non-linear models. For the ?rst controller a linear single-track model is used. This is a simpli?ed model of the vehicle where both the front and rear tires are assumed to be one lumped mass. With this linear model a PID controller is constructed. The LP provides the left and right rear wheel torques, from the yaw moment of the controller. The LP also takes into account the friction and engine limitations. This control algorithm is only compatible with a rear wheel drive vehicle. The second controller is based upon a non-linear two-track model. In this model each wheel is separately modeled. This means that the model is more accurate but more knowledge of the vehicle parameters is needed. With this two-track model it is possible to determine the torques for all four wheels. This means that this controller will be compatible with a four wheel drive vehicle. The LP controller is a combination of the PID and LP from the previous controller. The LP controller uses the equation of motions of the two-track model to determine the torques for the left and right rear wheels, with respect to the friction and engine limits. Both controllers show improvement of the lateral vehicle dynamics, compared to the dynamics of a vehicle with no torque vectoring. With torque vectoring the lateral acceleration is higher and therefore the velocity through corners is higher. These higher velocities through corners will translate to faster lap-times, which is the goal of this thesis. The second controller shows similar results with the ?rst controller, but seems to have a slight more aggressive control output. Both controllers determine only the left and right rear wheel torques. The potential of four wheel drive for the second controller is not simulated, as the Forze VI is only rear wheel drive. The LP controller can in the future use it's fully potential when Forze decides to build a four wheel drive vehicle.

torque vectoring

http://resolver.tudelft.nl/uuid:720ca646-2936-4680-9a63-b7665822c83f

Student theses

master thesis

(c) 2014 Stoop, A.W.