Vehicle Torque
Vectoring
Control
                                             ECE 1635
                                           April 6, 2015
     Christopher Au   Moeed Siddiqui   Yujie Guo
Agenda
                    Background
                            Plant
                      Controller
             Simulation Results
April 2015   Modern Control - Vehicle Torque Vectoring   2
Background
● Two types of undesirable vehicle steering dynamics
  ○ Understeer
  ○ Oversteer
● TV Advantages:
  ○ Improved handling
  ○ Traction when turning
  ○ Better overall performance in
     poor road conditions
April 2015           Modern Control - Vehicle Torque Vectoring   3
Plant Model
April 2015    Modern Control - Vehicle Torque Vectoring   4
Plant: Mathematical Models
 Reference Model:
 Vehicle Motion Model:
April 2015          Modern Control - Vehicle Torque Vectoring   5
Plant: Simulation Parameters
April 2015   Modern Control - Vehicle Torque Vectoring   6
Controller - State Feedback
Full State Feedback Controller
With Integral Action:
●     Controllable system
●     Pole placement using Matlab
Controllability Matrix:
Control Law:
Closed Loop System:
                                                       Block Diagram for Full State
                                                        Feedback Controller /with
                                                             Integral Action
 April 2015                Modern Control - Vehicle Torque Vectoring                  7
Controller - State Feedback
Tuning Full State Feedback Controller
With Integral Action:
Tuning Parameters:
              Step Response:                          Close Loop Bode Diagram:
 April 2015                    Modern Control - Vehicle Torque Vectoring         8
Controller - Sliding Mode
Sliding Mode Controller
● Discontinuous control signal
● Adds robustness to the closed-loop system
 Smoothed Error:
Control Law:
 April 2015          Modern Control - Vehicle Torque Vectoring   9
Controller - Sliding Mode
Consider the Lyapunov candidate function:
 Choose design parameter:
 April 2015           Modern Control - Vehicle Torque Vectoring   10
Simulation - HIL Setup
● HIL DEMO
             Steering Input                                         Vehicle Yaw Rate
April 2015                    Modern Control - Vehicle Torque Vectoring                11
Simulation - HIL Problems
● To resolve controller instability when using HIL:
  ○ Increased sampling period in Labview
  ○ Eliminated dead zone when motor changes direction
  ○ Added scaling to PD controller to replicate gearing
  ○ More aggressive LPF
             Motor speed PD controller
April 2015                Modern Control - Vehicle Torque Vectoring   12
Simulation Results - State Feedback
● State Feedback Controller Performance
  ○ approximate 0 steady state error
  ○ 1 sec delay during transients
  ○ maximum torque range -400N/m to +400N/m
      Steering Input                  Yaw Rate                     Torque Transfer
April 2015             Modern Control - Vehicle Torque Vectoring                     13
Simulation Results - Sliding Mode Control
● 0% ss error
● 0.5 second delay
April 2015           Modern Control - Vehicle Torque Vectoring   14
Controller Comparison
April 2015   Modern Control - Vehicle Torque Vectoring   15
Controller Simulation Video
April 2015    Modern Control - Vehicle Torque Vectoring   16
3DOF Bicycle Model
Distance and Angle Matrix:
Velocity Matrix:
April 2015         Modern Control - Vehicle Torque Vectoring   17
Conclusion
● Two controllers were design to implement torque
  vectoring
  ○ State feedback based on an augmented plant
  ○ Nonlinear sliding mode controller
● HIL simulation in Labview
  ○ Results show that sliding mode performs better
● Recommendations
  ○ Kalman Filter
  ○ Feedforward controller
  ○ Adaptive controller
April 2015         Modern Control - Vehicle Torque Vectoring   18
Thank You
                 Questions?
April 2015   Modern Control - Vehicle Torque Vectoring   19
References
[1] DSC CONTROL. (2012). Retrieved March 30, 2015, from
http://madstyle1972.com/MAZDA6_2014/servicehighlights/books/n6w04/html/id041500103900.html
[2] Burgess, M. Torque vectoring. Retrieved March 17, 2015, from
 http://www.vehicledynamicsinternational.com/downloads/VDI_Lotus_Vector.pdf
[3] NAGAI, M., HIRANO, Y., & YAMANAKA, S. (2007). Integrated Control of Active Rear Wheel Steering and Direct Yaw Moment
Control. Retrieved March 17, 2015, from
http://www.tandfonline.com/doi/abs/10.1080/00423119708969336#.VRlpCpPF8WU
[4] Aircraft Pitch: State-Space Methods for Controller Design. (2012). Retrieved March 17, 2015, from
http://ctms.engin.umich.edu/CTMS/index.php?example=AircraftPitch§ion=ControlStateSpace
[5] Slotine, J., & Li, W. (1990). Applied Nonlinear Control Paperback. Prentice Hall; 1 edition. Retrieved March 17, 2015, from
ftp://222.18.54.49/xiaomagecc/Applied%20Nonliear%20control%20[Slotin%201991--Prentice%20Hall].pdf
[6] Thang Truong, D., Meywerk, M., & Tomaske, W. (2013). Torque Vectoring for Rear Axle using Adaptive Sliding Mode Control.
Retrieved March 17, 2015, from
https://www.deepdyve.com/lp/institute-of-electrical-and-electronics-engineers/torque-vectoring-for-rear-axle-using-adaptive-sliding-
mode-control-4RQOOh9G9i
  April 2015                                  Modern Control - Vehicle Torque Vectoring                                            20