Predictive Traction Control

advertisement
Dynamic Traction
Control
By: Thiago Avila, Mike Sinclair & Jeffrey McLarty

Drastically improve vehicle performance and safety by
maintaining optimal wheel traction in all road conditions
Motivation
Acceleration
10.000
Acceleration [m/s2]
5.000
0.000
0
1
2
3
4
5
6
7
8
9
Centre of Gravity
Front Tire
Rear Right Tire
-5.000
Rear Left Tire
-10.000
-15.000
Motivation
Time [s]

FSAE car is currently traction limited and
would benefit from the use of a traction
control system

System must follow FSAE guidelines

Minimal cost solution should be pursued
Needs Assessment
◦ Meet FSAE Guidelines
◦ Predict slip with enough time to adjust engine
output
◦ Reduced FSAE 75m acceleration times
◦ Improve FSAE skid pad testing results
Design Criteria and Constraints

The traction control system is required to
prevent driver error from overloading any
of the four wheels and causing slip,
through either throttle or brake
application
Problem Formulation

Physics model sensors
◦ 3-axis Accelerometer
◦ Linear Potentiometer

Cost & Complexity
Engine Power Control
◦
◦
◦
◦
Cutting Spark
Difficult to Predict Power
Limiting Fuel
Improper Fuel Ratio
Drive by wire throttle
Infringes FSAE rules
Electronic Air Restrictor
Abstraction

Slip Model
◦ Vehicle Dynamics and Sensing

Vehicle Control
◦ Electronic Restrictor
Proposed Solution Breakdown

Slip Model
◦ Dynamic Physics Model
◦ Dynamic Coefficient of Friction
◦ Understeer Detection
Proposed Solution
External
Sensors
Slip
Angle
Radius
X/Y/Z
Acceleration
Driver
Pedal
Physics
Model
(Saturator)
RPM
Throttle Pos.
μs/μk
ECU
Design Layout
+
Wheel
Slip
Detector
CBR 600
F4i
Engine
Wheels
Physics Model
Torque Map
4500
3000
3500
4000
2500
Throttle Angle (Degrees)
90
67.5
45
22.5
2000
0
5500
5000
Engine Speed (RPM)
6000
6500
7000
7500
8000
8500
9000
9500
10000
10500
11000
11500
12000
12500
13000
13500
14000
20
10
0
Torque (N-m)
60
50
40
30
-10
Interpolate Between
Four Points on Torque
Map
•Interpolate between
Engine Speeds at
Throttle 1
Interpolation
Interpolate Between
Four Points on Torque
Map
•Interpolate between
Engine Speeds at
Throttle 1
•Interpolate between
Engine Speeds at
Throttle 2
Interpolation
Interpolate Between
Four Points on the
Torque Map
•Interpolate between
Engine Speeds at
Throttle 1
•Interpolate between
Engine Speeds at
Throttle 2
•Interpolate between
results at different
Throttles
Interpolation
Interpolate Between
Four Points on the
Torque Map
•Interpolate between
Engine Speeds at
Throttle 1
•Interpolate between
Engine Speeds at
Throttle 2
•Interpolate between
results at different
Throttles
Interpolation
Interpolate Between
Four Points on the
Torque Map
•Interpolate between
Engine Speeds at
Throttle 1
•Interpolate between
Engine Speeds at
Throttle 2
•Interpolate between
results at different
Throttles
•Engine Power from 4
point Interpolation =
Done
Interpolation
Physics Model

Installed Sensors
◦
◦
◦
◦
◦
◦
◦
◦
Steering Wheel Angle
2-D Acceleration
Suspension Deflection
Wheel Velocity
Brake Pressure
Engine RPM
Throttle Position
Air Mass Flow Rate
Data Acquisition
800
700
Normal Force (N)
600
500
Rear Left
400
Rear Right
FL Model
300
Front Right
200
100
0
0
25
50
75
100
125
150
175
200
225
250
275
300
325
350
375
Time
Physics Model Simulation
400
600
500
Vertical Force (N)
400
300
Modeled Vertical Force
Spring Force
200
100
0
0
25
50
75
100 125 150 175 200 225 250 275 300 325 350 375 400
Time
Model Validation – FL Tire
10.000
Acceleration [m/s2]
5.000
0.000
0
1
2
3
4
5
6
7
8
9
Centre of Gravity
Front Tire
Rear Right Tire
-5.000
Rear Left Tire
-10.000
-15.000
Time [s]
Slip
[True/False]
1
Slip
0
0
-1
1
2
3
4
5
Time [s]
Slip Condition
6
7
8
9
Calculate
Engine
Torque @
T(0)
Slip
Detected
Calculate
Vertical
Force @
T(0)
Calculate Coefficient of Friction and Update
Model
μs
Dynamic Coefficient of Friction
Calculator
Maintain
current
μs
Yes
No
No Slip
Detected
Increase
μs
Is μs at
the limit?
1.4
Initial Value
Coefficient of Friction
1.3
1.2
New Limit
1.1
1
0.9
0.8
0.7
Calculated Values
0.6
0.5
0
20
40
60
Time
80
100
Optimize Performance
120
140

Turning Radius:
◦ Desired vs. Actual

Major Factor:
◦ Wheel Slip Angle
Understeer Detection
Lateral Force vs. Slip Angle
1000
800
600
Lateral Force (lbf)
400
Goodyear7, 12 psi, IA0, load50
200
Goodyear7, 12 psi, IA0, load150
0
-15
-10
-5
0
5
10
Goodyear7, 12 psi, IA0, load250
Goodyear7, 12 psi, IA0, load350
-200
Goodyear7, 12 psi, IA0, load450
-400
-600
-800
-1000
Slip Angle [degrees]
Slip Angle
15

Vehicle Control
◦ Electronic Restrictor
◦ Brake Pressure Controller
Proposed Solution
Electronic Restrictor
Electronic Restrictor
Electronic Restrictor
Electronic Restrictor
Electronic Restrictor
Electronic Restrictor
Electronic Restrictor
Rotary
Potentiometer
Servo
Gears
Butter
-FlyValve
Electronic Restrictor
Tpeak  0.1sec
Tsettle  0.5 sec
%O.S .  5%
esteadystate  0
K
P( s) 
s( s   )
G ( s  p)
C ( s) 
( s  a)( s  b)
Electronic Restrictor
External
Sensors
Slip
Angle
Radius
X/Y/Z
Acceleration
Driver
Pedal
Physics
Model
(Saturator)
RPM
Throttle Pos.
ECU
Patents
μs/μk
+
Wheel
Slip
Detector
CBR 600
F4i
Engine
Wheels
External
Sensors
Slip
Angle
Radius
X/Y/Z
Acceleration
Driver
Pedal
Physics
Model
(Saturator)
RPM
Throttle Pos.
ECU
μs/μk
+
CBR 600
F4i
Engine
Wheels
Wheel
Slip
Detector
Possibly patentable:
Patents
Continuously Improving
Predictive Traction Control
Start
Order Parts
& Materials
1 day
Finish
Test &
Optimize
4 weeks
Program PSoC with
Physics Model &
Interpolation
3.5 weeks
Build
Restrictor
Install
Restrictor
2 weeks
1 week
Create Controller based on
Design Criterion
2.5 weeks
Commissioning
The Plan
Critical Path ~10 weeks
Questions?
Comments?
The End
Thank you!
Download