Control for

Maarten Steinbuch
Dept. Mechanical Engineering
Control Systems Technology Group
TU/e

Control for Dummies 1

Motion Systems

F
m

Control for Dummies 2

1
1. Introduction
2. Timedomain tuning
3. Frequency domain & stability
4. Filters
5. Feedforward
6. Servo-oriented design of mechanical systems

Control for Dummies 3

2. Time Domain Tuning

Control for Dummies 4

2
 x

Disturbance F
Servo force ? d
F Mass
s M

Control for Dummies 5

Mechanical solution:

x
k
F Disturbance Fd
Mass
M
d

Force spring- damper: F = k x d x&

1 k
Eigenfrequ
ency: f=
2 M

Control for Dummies 6

3
Servo analogon:

F
servo Mass
M

Servo Force: Fs = k p x kv x&

1 kp k p : servo stiffness
Eigenfrequency: f =
2 M kv : servo damping

Control for Dummies 7

Example:
x

Disturbance F
d
F Mass
s M

Slide: mass = 5 kg
Required accuracy 10 m at all times
Disturbance (f.e. friction) = 3 N

1. Required servo stiffness?

2. Eigenfrequency?

Control for Dummies 8

4
How to move to / follow a setpoint:

x
h or x
s

Spring- damper: F = k(h x) + d (h& x&)

Controller: Fs = k p (xs x) + kv (x&s x&)

Control for Dummies 9

x
xs
Fd

Fdisturbance

controller process
xs +
e d F x
+
k p + kv Mass
-
dt +
Fservo

Kp/kv-controller or PD-controller

Control for Dummies 10

5
 controller
xs
+ e x
kp = ? process
-
kv = ? Fservo

kp
error stability
kv
Trade off

Control for Dummies 11

Concluding remarks time domain tuning

A control system, consisting of only a single mass m and a

kp/kv controller (as depicted below), is always stable.
kp will act as a spring; kv will act as a damper
As a result of this: when a control system is unstable, it
cannot be a pure single mass + kp/kv controller
(With positive parameters m, kp and kv)

x
xs
kp
M
kv

Control for Dummies 12

6
 Setpoints:
x
xs

What should xs look like as a function of time, when moving the mass?
(first order, second order, third order,.?)

Control for Dummies 13

x
Apply a force F (step profile):

F (t ) = M&x&(t ) F M

x(t) is second order, when F constant

Second order profile requires following information:

- maximum acceleration
- maximum velocity
- travel distance

Control for Dummies 14

7
 Example

Pos = 2 6.3rad
Velmax = 20 63rad / sec
Acc max = 500 1.6e3rad / sec 2

Control for Dummies 15

3 Frequency domain

Time domain:
Monday and Thursday at 22:10

Frequency domain:
twice a week

Control for Dummies 16

8
 x

F M

Frequency Response Function H

amplitude (log)

x -2
|H| =
F

frequency (log)
phase
H
o
-180

Control for Dummies 17

going from Time-domain to the Frequency-domain

x excitation force in N and displacement in m
0.2
excitation force (offset 0.1 &scaling 1e-4)

0.15
F
M
0.1
weak spring M = 5 kg
(f = 2.5 Hz)
0.05

F (t ) = 400 sin(2 7t ) 0

-0.05
response
-0.1

-0.15
H (7 Hz ) 0.045 / 400 = 1e 4 m / N 0 0.5 1 1.5 2 2.5
time in sec
H (7 Hz ) 180 0

Control for Dummies 18

9
finding a solution of the equation of motion: x

F=M x F M
choose input:

F = F sin(t )
then:


x = x sin(t + ) x = ?; = ?
solution:

x(t ) = F 2 sin(t ) + c1t + c2

M
x 1
H= =
F M 2 H = x = 1 2
F M
log( H ) = log 1 2 log H = = 180
M

Control for Dummies 19

measurement mechanics stage

amplitude in dB

-20

-40

-60

-80
2 3
10 10
frequency in Hz
phase in deg
200

-200 2 3
10 10

Control for Dummies 20

10
 Derivation of transfer function

1. make a model of the dynamics: differential equations

2. substitute s=d./dt

3. rearrange the equations and get the transfer function e.g. H(s)

4. for sinusoids make a Bode plot using s=j

Control for Dummies 21

Transfer function:

x( s) 1
H (s) = = 2
F ( s ) Ms + ds + k

F 1 x
Ms 2 + ds + k

Control for Dummies 22

11
 consider sinusoidal signals ('Euler notation'):

x (t ) = x (cos t + j sin t ) = xe jt

x& (t ) = x ( sin t + j cos t ) = j xe jt

apparently: s = j for sinusoidal signals

Frequency Response Function:

s j
1
H ( j ) =
M 2 + jd + k

Control for Dummies 23

e d F
k p + kv
dt

F = k p e + k v e&

F ( s) = ( k p + k v s) e ( s)

transfer function:

F
C ( s) = ( s) = ( k p + k v s)
e

frequency response:

C = k p + jk v

Control for Dummies 24

12
 Amplitude: C = k p2 + kv2 2

0 C kp
C 0

C kv
C 90

log (| C |) = log kv + log

break point:
log k p = log k v + log
kp
=
kv

Control for Dummies 25

Bode plot of the PD-controller:

kp = 1500 N/m; kv = 20 Ns/m

amplitude in dB
100

80
+1
+0
60 0 1 2
10 10 10
frequency in Hz
phase in deg
100

50

0 0 1 2
10 10 10

Control for Dummies 26

13
 Block manipulation

e x
C(s) H(s) x ( s)
= C ( s) H ( s)
e ( s)

C +1 H -2
1
kp M

kp frequency
kv

-2
CH
-1
frequency

Control for Dummies 27

xs + e x
C(s) H(s)
-

x CH
Hc = =
xs 1 + CH

Control for Dummies 28

14
Four important transfer functions Fd
xs + e + x
C(s) H(s)
1. open loop: - Fs+
Ho ( s) = C( s) H ( s)

2. closed loop:
x C (s) H (s)
H c (s) = (s) =
xs 1 + C (s) H (s)

3. sensitivity:
e 1
S ( s) = ( s) =
xs 1 + C ( s) H ( s)

4. process sensitivity:
x H ( s)
H ps ( s) = ( s) =
Fd 1 + C ( s ) H ( s)

Control for Dummies 29

Fd
xs + e + x
C(s) H(s)
- Fs+

Derivation of closed-loop transfer functions:

start with the output variable of interest
go back in the loop, against the signal flow
write down the relations, using intermediate variables
stop when arrived at the relevant input variable
eliminate the intermediate variables

Control for Dummies 30

15
 amplitude in dB
closed loop
20

-20

-40
2 3
phase in deg 10 10
200

Experimental results:
-200 2 3
10 10
stage servo
amplitude in dB open loop
20

-20

-40
2 3
10 frequency in Hz 10
phase in deg
200

-200 2 3
10 10
Control for Dummies 31

bandwidth: 0 dB crossing open loop

(cross-over frequency)

Control for Dummies 32

16
 The Nyquist curve

amplitude in dB Bode plot Nyquist plot

50 1.5

0 1

-50
0.5
amplitude (appr 0.7)
-100 1 2
10 10
0
phase in deg frequency
0 phase (appr. -175 deg)
-0.5

-100
-1

-200 -1.5
1 2
10 10 -1.5 -1 -0.5 0 0.5 1 1.5
frequency

Control for Dummies 33

Stability:
The open-loop FRF CH(j) should have the (-1,0) point at left side
Nyquist plot
1.5
Im

0.5

Re
0

-0.5

-1

-1.5
-1.5 -1 -0.5 0 0.5 1 1.5

Control for Dummies 34

17
 4. Filters Mimo

Integral action
Differential action PeeDee
Low-pass PeeEye
High-pass
Band-pass
Notch (sper) filter

Control for Dummies 35

Integral action :

1 Y(t)
X(t)
is

I integral time constant I =1/ki

-1

0 =2f

-90
Control for Dummies 36

18
 Differential action
+1

u u
H = ks = ; s = j; = k
+90

u 0
ks

1
= d = 2fd
ks u d
tamme differentiator :=
d s +1
+90
0

Control for Dummies 37

lead filter
u 1 + 1s 1 + d s
H= = = 1
1 + 2s 1 + d s 25
2

>1 1
1 c = 1 2 =
1 2

+90

0
1 1
1 = 2 =
1 2
Control for Dummies 38

19
 P+I+D

u 1 1 + d s
H = = k 1 +
i s 1 + d s

-1 +1

1 1
i = d =
i d
+90

-90

Control for Dummies 39

2nd order filter H=

u
=
k
2
s s
2
+ 2 +1
1 1 1 1
(t) u(t)
- s s

2 1 Top: o = 1 1 2 .
2
1

-2
0
-90
-180
= 1

Control for Dummies 40

20
 General 2nd order filters

2
General: 12 2

1
s2 s
2
+ 2 1 +1 +2
u 1 1
H= = 2 1
s s
+ 2 2 +1
22 2

+180

Control for Dummies 41

1 2 +2

22 -2
2

1

-180
2 1
Control for Dummies 42

21
 Notch-filter :1= 2

ampl.
1
2

fase 0

-180
Control for Dummies 43

W.B.E.

Control for Dummies 44

22
 Loop shaping procedure

1. stabilize the plant:

add lead/lag with zero = bandwidth/3 and pole =
bandwidth*3, adjust gain to get stability; or add
a pure PD with break point at the bandwidth
2. add low-pass filter:
choose poles = bandwidth*6
3. add notch if necessary, or apply any other kind
of first or second order filter and shape the loop
4. add integral action:
choose zero = bandwidth/5
5. increase bandwidth:
increase gain and zero/poles of integral action,
lead/lag and other filters

during steps 2-5: check all relevant transfer

functions, and relate to disturbance spectrum

Control for Dummies 45

Implementation issues

1. sampling = delay: linear phase lag

for example: sampling at 4 kHz gives phase lag

due to Zero-Order-Hold of:

180 @ 4 kHz
18 @ 400 Hz
9 @ 200 Hz

2. Delay due to calculations

3. Quantization (sensors, digital representation)

Control for Dummies 46

23
 5 Feedforward design

Control for Dummies 47

Why feedforward?

Consider the simple motion system

1

F
m
Setpoint x s

x
0
0 t [s] 1.8

Control problem: track setpoint xs

Is this possible with a PD-controller?

Control for Dummies 48

24
 Analysis (IV)
-3
x 10
2

m = 5 [kg]
1.5
K v= 260 [Ns/m]
1

K p= 6500 [N/m]
error [m]

0.5

-0.5 K p= 65000 [N/m]

-1

-1.5

-2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2

t [s]

Control for Dummies 49

Feedforward based on inverse model

ms 2

xs 1 x
K p + Kv s
ms 2

Control for Dummies 50

25
Example: m=5 [kg], b=1 [Ns/m], 2nd degree setpoint
1

xs [m] 0.5

1.5
vs [ms-1]

0.5

4
as [ms-2]

-2

-4
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
t [s]

Control for Dummies 51

Example: tracking error, no feedforward

-3
x 10
2

1.5

1 viscous damping effect

0.5
error [m]

-0.5

-1

-1.5

-2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
t [s]

Control for Dummies 52

26
 Example: tracking error, with feedforward
-3
x 10
2

1.5 K fv= 0.9, K fa= 0

1

0.5
K fv= 0.9, K fa= 4.5
error [m]

-0.5

-1

-1.5

-2
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
t [s]

Control for Dummies 53

feedforward structure

sign( x& s )
K fc

x&&s
K fa

x& s
K fv

xs x
C(s) H(s)

Control for Dummies 54

27
 3rd degree setpoint trajectory
1.5

xs [m]
1

0.5

1
vs [ms-1]

0.5

4
as [ms-2]

-2

-4
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8
t [s]

Control for Dummies 55

6. Servo-oriented design of
mechanical systems

Control for Dummies 56

28
Example of measurement:
mechanical system (force to position)

modelling understanding the dynamical

behaviour
Control for Dummies 57

Three Types of Dynamic Effects

- Actuator flexibility

- Guidance flexibility

- Limited mass and stiffness of frame

Control for Dummies 58

29
1. Actuator flexibility

k
F
s Motor Sensor
d

Control for Dummies 59

2. Guidance flexibility

F M,
s J
k

Control for Dummies 60

30
3. Limited mass and stiffness of frame

F
s
Motor

Frame

Control for Dummies 61

M1 M2

Positioning the load M2 (while using x1 for feedback):

Rule of thumb:
Optimal bandwidth with 0 dB crossing of open loop between the
antiresonance and resonance frequency of the mechanical system.

Control for Dummies 62

31
Concluding Remarks

bit of control into mechanical design

bit of mechanics into control design
same language (mechatronics)

Control for Dummies 63

32