Fuzzy Automated Braking System

For Collision Prevention

by
Piero P. Bonissone & Kareem Aggour
 Agenda

! Problem Description (Objective,Assumptions)

! Related Work
! System Design
! Dynamic Simulator
! Fuzzy Controller Design
! Genetic Algorithm Tuning

! Tuning Results
! Analysis & Conclusion
2
 Motivation
! Driver reaction time often delays
identification of potentially hazardous
situations

! Many such accidents could be prevented

3
 Objective

! Use an embedded Fuzzy Logic controller

Whew!

! To automatically identify and react to

potentially hazardous driving situations

4
 Related Work on FC for Braking

! US Patent 5634698 (Cao et al. of Germany):

! System for Controlling Brake Pressure Based on
Fuzzy Logic Using Steering Angle and Yaw Speed
! This system is focused more on reducing vehicle
skidding than on reducing the vehicle speed to
prevent collisions
! US Patent 5416709 (Yeh et al. of Taiwan):
! Fuzzy Controller for Anti-Skid Brake Systems
! This system is similar to Caos system and is also only
vaguely similar to the goal of the second fuzzy
controller designed.
5
 Model Assumptions
! Straight line on a surface with no gradient
! Controlled vehicle's wheels do not slip
! Limits the simulation to consider only static friction,
disregarding the kinetic friction generated by wheel
slip
! Linear correlation between the application of
brake pressure and the deceleration the vehicle
experiences.
! If 25% of the maximum possible brake pressure is
applied, then the vehicle will decelerate by 25% of its
maximum possible deceleration.
6
 Sensor & Actuator Assumptions

! The sensor technology required to measure

the state variables used by the proposed
system (range, relative velocity, and friction
coefficient) is readily available.
! Laser range finders, and Doppler radars could
provide the first two measurements,
! Wheel-slip sensors currently used for automobile
Traction Control Systems or Anti-lock Breaking
Systems (ABS) could estimate the third state
variable.
! The required actuators are already installed in
all automobiles equipped with ABS.
7
 Variables of Interest
! Separation between the two vehicles
L

! Differences in their velocities

VTracking VTracked

V = Vtracked (t ) Vtracking (t )
! Surface conditions (i.e., friction coefficient
between the vehicle tires and the road)

8
 Simulator-FC-GA Architecture
Final Separation Distance
CARS SIMULATION

Difference Percentage of
in velocity brake pressure

GENETIC ALGORITHMS Separation FUZZY CONTROLLER(S)

Distance

KB

Scaling Factors

(Normalized) Term Sets

Rule Set
9
 Tuning FC Knowledge Base with GA
! A Mamdani- type FLC approximates a relationship
between a state X and an output Y by using a KB
and a reasoning mechanism (interpreter)
implemented by the generalized modus-ponens.
! The Knowledge Base (KB) is defined by:

Scaling Factors (SF):

Parameters

! Ranges of values of state and output variables

! (Normalized) Term Set (TS):
! Membership functions of values, normalized by SF
Structure

! Rule Set (RS):

! Syntactic mapping of symbols from X to Y
10
 FC Sensitivity to Parameter
and Structure Changes
X1 X2
Very Low Low Medium High Very High
Very Low PH PH PM PL ZE
Changing Low PH PM PL ZE NL
a Scaling Medium PM PL ZE NL NM
High PL ZE NL NM NH
Factor Very High ZE NL NM NH NH

X1 X2
Very Low Low Medium High Very High
Very Low PH PH PM PL ZE
Changing Low PH PM PL ZE NL
a Term in Medium PM PL ZE NL NM
High PL ZE NL NM NH
X1 Very High ZE NL NM NH NH

X1 X2
Very Low Low Medium High Very High
Very Low PH PH PM PL ZE
Changing Low PH PM PL ZE NL
a Rule Medium PM PL ZE NL NM
High PL ZE NL NM NH
Very High ZE NL NM NH NH
 FC Sensitivity to Parameter
and Structure Changes
X1 X2
Very Low Low Medium High Very High We use Non-
Very Low PH PH PM PL ZE
Changing Low PH PM PL ZE NL normalized
a SF Medium PM PL ZE NL NM
Term Sets, e,g,
High PL ZE NL NM NH
Very High ZE NL NM NH NH
the SF are
X1 X2
incorporated
Changing Very Low Low Medium High Very High
into the
a Term Very Low PH PH PM PL ZE

in a TS
Low
Medium
PH
PM
PM
PL
PL
ZE
ZE
NL
NL
NM
Term Sets.
High PL ZE NL NM NH
(e.g. X1) Very High ZE NL NM NH NH

So we will
X1
Very Low Low
X2
Medium High Very High
Focus our
Changing Very Low PH PH PM PL ZE tuning to the
a Rule Low PH PM PL ZE NL

in RS
Medium PM PL ZE NL NM Term Sets
High PL ZE NL NM NH
Very High ZE NL NM NH NH
 Simulator-FC-GA Architecture
CARS
FC1 & FC2 SIMULATION
Decoder
Termsets
Fitness
Function
Final Separation Distance
Percentage
Percentage
Difference of brake
of brake
in velocity pressure
FC1(ideal FC2
pressure
Separation (actual
surface)
Distance surface)

Parents Offspring Mutation KB1 KB2

Selection by Friction
Crossover Coefficient
% Unaffected by FUZZY
Crossover (SSGA)
CONTROLLERS
Elitist
(best from Gen i)
Seven scenarios
Gen. Score P(selection) Gen.
(initial conditions for:
(i) (i+1)
GENETIC L, Vtracked,Vtracking,
ALGORITHMS VDOTtracked )
13
 Simulator Design Equations

Vo + V f
d= t V f = Vo + at
2
1
momentum= mV 2
Workfriction = mgd
2
Work net = Momentum
1 1
Work friction = mV f
2
mV o
2

2 2
1
mgd = mV o
2

2
14
 Simulator Design Equations

Vo
2 2
V
d stopping = o
a =
2g 2d

a maximum = g

V ( t + 1 ) = V ( t ) + ( a maximum % Braking ) t

d separation (t + 1) = d separation (t ) + (Vtracked (t + 1) Vtracking (t + 1))t

15
 Fuzzy Controller(s) Design
! Designed system containing two fuzzy controllers
(FC) connected in series
Difference in
velocity % Pressure
(Ideal surface)
Separation Fuzzy Controller 1
% Pressure
distance
Fuzzy Controller 2 (Actual surface)
s

! Each FC Interpreter uses:

! LHS evaluation and rule Firing: Minimum Operator
! Rule Aggregation: Maximum Operator
! Defuzzification Method: Center of Gravity
! Each KB defined by:
! (Non-normalized) Term Sets & Rule Set 16
 Rule Sets FC 1
Separation Distance (L)Term Set Term Sets
Velocity Term Set

Control Output: %PIdeal

VLL LL ML HL VHL F = Full
VH = Very High
VLV M%P L%P VL%P Z%P Z%P H = High
M = Medium
LV H%P M%P L%P VL%P Z%P VL = Very Low
Z = Zero
MV VH%P H%P M%P L%P VL%P
State Variables: V, L
VL = Very Low
HV F%P VH%P H%P M%P L%P L = Low
M = Medium
VHV F%P F%P VH%P H%P M%P H = High
VH = Very High

17
 Rule Sets FC 2
Coefficient of Friction Term Set Term Sets
% Brake Pressure T. Set

VL L M H VH Control Output %PActual

F = Full
Z%P Z%P Z%P Z%P Z%P Z%P VH = Very High
H = High
VL%P H%P M%P L%P VL%P VL%P M = Medium
VL = Very Low
L%P VH%P H%P M%P L%P L%P Z = Zero

State Variables:
M%P F%P VH%P H%P M%P M%P % PIdeal

H%P F%P F%P VH%P H%P H%P F = Full VL = Very Low

VH = Very High L = Low
VH%P F%P F%P F%P VH%P VH%P H = High M = Medium
M = Medium H = High
F%P F%P F%P F%P F%P F%P VL = Very Low VH = Very High
Z = Zero

18
 Genetic Algorithm
Tuning Term Sets
! Chromosome Design:
! Parametric representation of the term sets:
! Slopes intersection at 0.5, and core (1 side)

core core core

1
membership

(Generic
Term Set)
0

! Guarantee that for each point in the universe of discourse the sum
of its associated membership values equal one
! Reduces degrees of freedom (by )
! Preserved a good property of the term set, which usually provides
good interpolation behavior
19
 GA Parameters
! Encoding = real-valued ! Elitist strategy = YES
! Population size = 200 ! Crossover type = single cut
! Probability Mutation = 0.5% ! Maximum number generations
! Probability Crossover = 30% = 2,000
! Fitness Scenarios = 15 ! Selection type = Rank

Fitness Function:
Final <= (-10, (-7.5, (-5, (-2, (0, (1, (3.5, (5, (7, (10, 12
(2.5, 3.5]
Separation -10 -7.5] -5] -2] 0] 1] 2.5] 5] 7] 10] 12] <=
Reward -25 -20 -15 -10 -5 5 8 15 10 5 2 -2 -5

20
 Tuned Term Sets
1.2 1.2

Difference in Velocity (V) Separation Distance

1 1

VL L M H VH VL L M H VH
0.8 0.8

Membership
Membership

0.6 0.6

0.4 0.4

0.2 0.2

0 0
0 10 20 30 40 50 60 70 0 20 40 60 80 100

Velocity (km/h) Distance (meters)

1.2
1.2

Coefficient of Friction % Brake Pressure

1
1

VL L M H VH Z VL L M H VH F
0.8
0.8
Membership
Membership

0.6
0.6

0.4
0.4

0.2
0.2

21
0 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Brake Pressure (%)
Coefficient of Friction (mu)
 Tuned Term Sets (Friction Coefficients)
1.2

Coefficient of Friction

VL L M H VH Typical Friction Coefficients

[from Forensic Dynamics, Inc]
0.8

0.80 = Dry Concrete (VH)

M embership

0.6 0.70 = Dry Asphalt (VH)

0.55 = Wet Concrete (H)
0.4
0.45 = Wet Asphalt (M)
0.30 = Snow (L)
0.15 = Ice (VL)
0.2

0
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Ice
Coefficient of Friction (mu) Dry Concrete
Snow Dry Asphalt
Wet Asphalt Wet Concrete 22
 Analysis & Conclusion
! Performance:
! Tested against hundreds of random scenarios with wide
range of initial distances, velocity and friction coefficients
! Consistently stopped within 3m ( 0.5m) of the tracked vehicle
(when physically possible)
! Occasional accidents were possible only when tracked vehicle
defied laws of physics due to unrealistic random initial conditions
! Two controllers in series worked well
! Reduced size of rule sets by ~ 50%
! # Rules in cascading controllers: 5X5 + 7x5 = 60
! # Rules in a 3-input controller: 5X5X5 = 125
! The GA tuning of the Term Sets was very successful and
considerably improved performance of the manually
designed FCs
23
 Comments for future Work
! L could be a function of Delta V
! We could add a second goal of tracking
vehicle at safe distance

24