Vehicle handling control using active differentials

This thesis describes an investigation into vehicle handling control using active differentials in the rear axle of a motor vehicle. Such devices are able to transfer torque between the rear wheels and have traditionally been used to improve traction whilst minimising the impact on vehicle handling. However, the capacity to generate a lateral torque difference across an axle also gives them the potential to be used for yaw moment control. In order to generate a rigorous assessment of this potential, the investigation is carried out in three distinct phases. Firstly, an analysis of the scope for modifying vehicle handling given unrestricted control over torque transfer between the rear wheels is carded out in the simulation environment. For this purpose an idealised yaw sideslip controller is developed. This is used to show that an ideal active differential can have significant yaw moment authority in terms of generating both understeer and oversteer and that this can be used to actively modify a vehicle's handling balance and apply stability control at the limits of adhesion. In the second phase, the capabilities of two types of contemporary active differential, the torque vectoring differential (TVD) and active limited slip differential (ALSID), are then assessed against the ideal differential and against a brake based yaw moment controller. TVDs are found to be able to offer very similar performance to both their ideal counterpart and to the brake based system. They Gan also deliver this performance with a fraction of the energy loss that is observed in the brakes, thus making TVDs a viable proposition for applying continuous yaw control below the limits of adhesion. ALSDs, on the other hand do not offer equivalent functionality to an ideal active differential but are still shown to be very effective stability control devices. In the third phase, the ALSID results are validated on a prototype vehicle where it is shown that they do indeed offer substantial stability improvements both on high and low-P surfaces. However in order to deliver such benefits and be practical for implementation, it is also shown that significant redevelopment of the idealised controller is required. Finally, with the ALSID operating alongside a commercial brake based stability control system, it is proven that substantial reductions in brake intervention can be achieved without significant controller integration.