Riassunto analitico
The aim of this work is to create a tire model for low adherence surfaces to be used on the driving simulator.
The starting point is a Pacejka MF-Tire 6.2 automatically fitted on the experimental measures made on a scraped ice surface.
In a second time, the same tire model must be finely tuned manually to fix possible interpolation problems and to adapt the model to different surfaces; this last process is made based on full vehicle measures.
The tire model is so scaled in a way that the Adams Car vehicle model shows the same behaviour of the measured real vehicle.
This process is made via some standardized maneuvers:
• Slow ramp steer: with the vehicle traveling at constant speed a steer slow ramp is applied in order to achieve a quasi-static maneuver.
• Frequency sweep steer: with the vehicle traveling at constant speed a sinusoidal steer signal, which frequency is progressively increasing in time, is applied to analyze the frequency response function of the full vehicle with the tires working in their linear range.
• Coast down: the deceleration is measured for the vehicle in coast down conditions to validate the tire rolling resistance.
• Straight line acceleration: an acceleration at the tire limit is performed with the vehicle starting standing still.
• Straight line braking: the vehicle traveling at a certain speed a full brake is applied in order to make the ABS work.
On the Adams Car vehicle it was decided not to perform the longitudinal validation due to the complex engine and ABS modeling with respect to CarRealTime.
Subsequently, the Adams model is translated into a CarRealTime one and the two vehicles are correlated with the same validation procedure; the possible differences are caused mainly by the vehicle model approximations introduced by CarRealTime.
Afterwards the model was validated on a track lap (via a max performance event on CarRealTime) on the ice lake handling track of the reference proving ground.
In a second moment, some sessions were performed at the driving simulator; here it is implemented an hardware in the loop system embedding the real car components. Thanks to this system, the vehicle braking behaviour could be validated obtaining a very good correlation while on the acceleration side smaller results were achieved since the traction control modeling is to improve in the HiL system.
The driving simulator sessions allowed us to collect very positive feedbacks from the drivers regarding the overall handling behaviour of the vehicle.
The future development should focus on implementing the traction control in the HiL system; this would allow to achieve a full validation.
Moreover, it should be performed another driving simulator session on the ice handling track to test the latest validation proposal: it was in fact modified after the last driving simulator session in order to correct the spotted miscorrelations.
|
Abstract
The aim of this work is to create a tire model for low adherence surfaces to be used on the driving simulator.
The starting point is a Pacejka MF-Tire 6.2 automatically fitted on the experimental measures made on a scraped ice surface.
In a second time, the same tire model must be finely tuned manually to fix possible interpolation problems and to adapt the model to different surfaces; this last process is made based on full vehicle measures.
The tire model is so scaled in a way that the Adams Car vehicle model shows the same behaviour of the measured real vehicle.
This process is made via some standardized maneuvers:
• Slow ramp steer: with the vehicle traveling at constant speed a steer slow ramp is applied in order to achieve a quasi-static maneuver.
• Frequency sweep steer: with the vehicle traveling at constant speed a sinusoidal steer signal, which frequency is progressively increasing in time, is applied to analyze the frequency response function of the full vehicle with the tires working in their linear range.
• Coast down: the deceleration is measured for the vehicle in coast down conditions to validate the tire rolling resistance.
• Straight line acceleration: an acceleration at the tire limit is performed with the vehicle starting standing still.
• Straight line braking: the vehicle traveling at a certain speed a full brake is applied in order to make the ABS work.
On the Adams Car vehicle it was decided not to perform the longitudinal validation due to the complex engine and ABS modeling with respect to CarRealTime.
Subsequently, the Adams model is translated into a CarRealTime one and the two vehicles are correlated with the same validation procedure; the possible differences are caused mainly by the vehicle model approximations introduced by CarRealTime.
Afterwards the model was validated on a track lap (via a max performance event on CarRealTime) on the ice lake handling track of the reference proving ground.
In a second moment, some sessions were performed at the driving simulator; here it is implemented an hardware in the loop system embedding the real car components. Thanks to this system, the vehicle braking behaviour could be validated obtaining a very good correlation while on the acceleration side smaller results were achieved since the traction control modeling is to improve in the HiL system.
The driving simulator sessions allowed us to collect very positive feedbacks from the drivers regarding the overall handling behaviour of the vehicle.
The future development should focus on implementing the traction control in the HiL system; this would allow to achieve a full validation.
Moreover, it should be performed another driving simulator session on the ice handling track to test the latest validation proposal: it was in fact modified after the last driving simulator session in order to correct the spotted miscorrelations.
|
