Development and Evaluation of a Torque-Vectoring Algorithm on RWD Racing Cars using a Dual Clutch
(2019)Department of Automatic Control
- Abstract
- Vehicle safety and vehicle performance are becoming more and more important for the society and many students, doctoral students, and researchers are interested in this field. Formula Student is a student competition that enables students to develop their own racing vehicles without any strict rules on how the vehicle should be controlled. The competition rules are instead directed into vehicle design and maneuverability on the track. This thesis was performed in collaboration with Lund Formula Student and BorgWarner. It presents a torque-vectoring algorithm that is planned to be implemented on a Formula Student car for next years competition in 2021. The Formula Student car will use a double clutch that is also developed at BorgWarner by... (More)
- Vehicle safety and vehicle performance are becoming more and more important for the society and many students, doctoral students, and researchers are interested in this field. Formula Student is a student competition that enables students to develop their own racing vehicles without any strict rules on how the vehicle should be controlled. The competition rules are instead directed into vehicle design and maneuverability on the track. This thesis was performed in collaboration with Lund Formula Student and BorgWarner. It presents a torque-vectoring algorithm that is planned to be implemented on a Formula Student car for next years competition in 2021. The Formula Student car will use a double clutch that is also developed at BorgWarner by a student in Mechanical Engineering at Lund University. The double clutch enables independent control of torques for each rear wheel.
The algorithm was developed in MATLAB/Simulink, mainly using vehicle models provided by BorgWarner. The goal with the torque-vectoring algorithm is to improve the vehicle’s accelerating behavior while cornering.
The nonlinear model that uses a torque-vectoring dual clutch (TVDC) is compared to another nonlinear vehicle model that represents a Formula Student vehicle using a limited slip differential (LSD) clutch. The controller is using yaw rate as a control signal. The results show that the vehicle trajectories for a lane change and U-turn coincide with the reference value, for the circular path, whereas the actual yaw rate value diverges from the reference value after some time. Overall, the vehicle can better follow the desired path with the proposed torque-vectoring algorithm for a double clutch than a vehicle using an LSD clutch. When the vehicle is accelerating,
it is clearly seen that for the TVDC, the actual yaw rate is following the desired yaw rate better than using open differential (OD) or LSD. Considering the yaw-rate error analysis, it is clearly seen that the error is smaller for a lane change and a U-turn than it is when the vehicle is following a circular path. Overall, the yaw-rate error is smaller when employing TVDC than without. (Less)
Please use this url to cite or link to this publication:
http://lup.lub.lu.se/student-papers/record/8987522
- author
- Grahovic, Mia and Rosicki, Madeleine
- supervisor
- organization
- year
- 2019
- type
- H3 - Professional qualifications (4 Years - )
- subject
- report number
- TFRT-6081
- ISSN
- 0280-5316
- language
- English
- id
- 8987522
- date added to LUP
- 2019-08-30 09:30:30
- date last changed
- 2019-08-30 09:30:30
@misc{8987522,
abstract = {{Vehicle safety and vehicle performance are becoming more and more important for the society and many students, doctoral students, and researchers are interested in this field. Formula Student is a student competition that enables students to develop their own racing vehicles without any strict rules on how the vehicle should be controlled. The competition rules are instead directed into vehicle design and maneuverability on the track. This thesis was performed in collaboration with Lund Formula Student and BorgWarner. It presents a torque-vectoring algorithm that is planned to be implemented on a Formula Student car for next years competition in 2021. The Formula Student car will use a double clutch that is also developed at BorgWarner by a student in Mechanical Engineering at Lund University. The double clutch enables independent control of torques for each rear wheel.
The algorithm was developed in MATLAB/Simulink, mainly using vehicle models provided by BorgWarner. The goal with the torque-vectoring algorithm is to improve the vehicle’s accelerating behavior while cornering.
The nonlinear model that uses a torque-vectoring dual clutch (TVDC) is compared to another nonlinear vehicle model that represents a Formula Student vehicle using a limited slip differential (LSD) clutch. The controller is using yaw rate as a control signal. The results show that the vehicle trajectories for a lane change and U-turn coincide with the reference value, for the circular path, whereas the actual yaw rate value diverges from the reference value after some time. Overall, the vehicle can better follow the desired path with the proposed torque-vectoring algorithm for a double clutch than a vehicle using an LSD clutch. When the vehicle is accelerating,
it is clearly seen that for the TVDC, the actual yaw rate is following the desired yaw rate better than using open differential (OD) or LSD. Considering the yaw-rate error analysis, it is clearly seen that the error is smaller for a lane change and a U-turn than it is when the vehicle is following a circular path. Overall, the yaw-rate error is smaller when employing TVDC than without.}},
author = {{Grahovic, Mia and Rosicki, Madeleine}},
issn = {{0280-5316}},
language = {{eng}},
note = {{Student Paper}},
title = {{Development and Evaluation of a Torque-Vectoring Algorithm on RWD Racing Cars using a Dual Clutch}},
year = {{2019}},
}