This master thesis was performed together with the Vehicle Dynamics department at Volvo Cars, Torslanda. Volvo Cars is a global producer of passenger cars. With increasing demand of safety, vehicle stability and efficiency, coupled to customer demands of agility and driving pleasure from passenger cars, there is a need of active systems to both enhance the vehicle dynamics while preserving or increase the efficiency of the vehicle. Therefore, it is of necessity to evaluate possible alternatives to existing systems in order to fulfil these demands. The scope of this Master Thesis was to evaluate the control algorithm “G-vectoring”, used to control yaw response in transient manoeuvres by controlling the longitudinal acceleration and thereby the load transfer. Further, the control algorithm was to be used with the ICE-driveline as actuator to obtain the longitudinal acceleration. The controller was to be evaluated in terms of handling, comfort, applicability and efficiency. The project consisted of two phases: first, a full factorial parameter study was performed using six vehicle specific factors and 1 environmental factor, friction coefficient, to evaluate the general effects of the driveline controller. Further, to understand on which vehicle the controller would be most applicable on in terms of handling and comfort, and how it affected the general efficiency of the vehicles. The full factorial test was done using simulations, using a one-track model extended with a simple load transfer model, and magic-formula tyres. The driveline used was a Volvo T5 FWD petrol driveline with 6 speed automatic Gearbox. The full factorial experiment was done using three manoeuvres, step steer, sine with dwell, and closed loop driving. The second phase was to answer how the controller affected the handling and comfort of the vehicle using subjective and objective assessment. In this stage, the controller was implemented to be used by real drivers in Volvo Cars driving simulator, using VI-CarRealTime as simulation tool. The controller was tuned for the nonlinear system with gain scheduling approach, and a linear description of the system was developed. The subjective assessment was done over a set of two tests, the first consisting of free driving in the linear range on a 3km test track for 8 laps, the first 4 with the controller active, and second with the controller inactive. The second was a set of 10 consecutive ISO 3888-2 Double Lane changes (DLC), the first 5 with the controller active, and the last 5 with the controller inactive. The subjective assessment was done with a 1-10 scale similar to SAE J1441. The objective evaluation was done only using the data obtained from the DLC test, as it poses a confined path for the driver to follow. The results from the full factorial test show that the controller can affect the yaw damping, yaw stability, steering angle variance and yaw response of a passenger car. Further, the results indicate that the controller can be expected to have the largest effect on a vehicle that has low mass, low inertia and a high centre of gravity. It was also found that the controller poses a more fuel-efficient way of controlling yaw compared to a ESC equipped vehicle. From the driving simulator test, it was found that the controller had better subjective rating overall in terms of steering and handling on a general steering feel and handling feel level. On a more detailed level the subjective assessment saw the largest advantages of the controller in terms of straight ahead response, under/oversteer at power on/off, and transitional stability, controllability and capacity feel. The disadvantage is primarily less torque feedback and less controllability at power on/off. In the objective metrics, the effect of the controller indicates a general decrease of the steering variance and an a decrease of the steering peak-to-peak values. The conclusion is that G-vectoring can enhance the lateral response and reduce the steering effort of a passenger car, and that the effects from G-vectoring can be felt by the driver.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:kth-309517 |
| Date | January 2017 |
| Creators | Hvitfeldt, Henrik |
| Publisher | KTH, Fordonsdynamik |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | TRITA-AVE, 1651-7660 ; 2017:77 |
Page generated in 0.0026 seconds