sensors

Article
Application of MEMS Accelerometers and
Gyroscopes in Fast Steering Mirror Control Systems
Jing Tian 1,2,3,4, *, Wenshu Yang 1,3 , Zhenming Peng 2 , Tao Tang 1,3 and Zhijun Li 1,3
1 Institute of Optics and Electronics, Chinese Academy of Science, Chengdu 610209, China;
yws@ioe.ac.cn (W.Y.); prettang@gmail.com (T.T.); zhijunhome@sina.com (Z.L.)
2 School of Opto-Electronic Information, University of Electronic Science and Technology of China,
Chengdu 610054, China; zmpeng@uestc.edu.cn
3 Key Laboratory of Optical Engineering, Chinese Academy of Sciences, Chengdu 610209, China
4 University of Chinese Academy of Sciences, Beijing 100039, China
* Correspondence: abb1978@163.com; Tel.: +86-189-8002-6386

Academic Editors: Slawomir Wiak and Manuel Pineda Sanchez

Received: 30 December 2015; Accepted: 22 March 2016; Published: 25 March 2016

Abstract: In a charge-coupled device (CCD)-based fast steering mirror (FSM) tracking control
system, high control bandwidth is the most effective way to enhance the closed-loop performance.
However, the control system usually suffers a great deal from mechanical resonances and time
delays induced by the low sampling rate of CCDs. To meet the requirements of high precision and
load restriction, fiber-optic gyroscopes (FOGs) are usually used in traditional FSM tracking control
systems. In recent years, the MEMS accelerometer and gyroscope are becoming smaller and lighter
and their performance have improved gradually, so that they can be used in a fast steering mirror
(FSM) to realize the stabilization of the line-of-sight (LOS) of the control system. Therefore, a tentative
approach to implement a CCD-based FSM tracking control system, which uses MEMS accelerometers
and gyroscopes as feedback components and contains an acceleration loop, a velocity loop and a
position loop, is proposed. The disturbance suppression of the proposed method is the product of
the error attenuation of the acceleration loop, the velocity loop and the position loop. Extensive
experimental results show that the MEMS accelerometers and gyroscopes can act the similar role as
the FOG with lower cost for stabilizing the LOS of the FSM tracking control system.

Keywords: MEMS accelerometer; MEMS gyroscope; acceleration feedback control; multi-loop control;
light of sight stabilization

1. Introduction
The fast steering mirror (FSM) control system is extensively applied in optoelectronic tracking
equipment, which is increasingly mounted on airplanes, vessels, vehicles and other moving platforms.
The traditional FSM control system usually uses fiber-optic gyroscopes (FOGs) and charge-coupled
devices (CCDs) to implement strap-down stabilization. The disturbance suppression of the FSM
control system is limited for its only two closed-loops [1–3].
In recent years, the noise of MEMS gyroscopes has been reduced with the development of research
and production in MEMS industry. The noise density of the SCR1100-D02 gyroscopes used in this paper
?
reaches 0.0085 (˝ /s){ Hz [4]. MEMS gyroscopes can be mounted on the frame of FSMs and used for
the stabilization of the FSM control system as they have relatively smaller size, lower weight and lower
power consumption. However, the bandwidth of MEMS gyroscopes with low noise is general less than
100 Hz, which limits the bandwidth of the velocity closed-loop and the disturbance suppression ability
of the FSM control system. In order to further improve the disturbance suppression ability of the
FSM control system, an acceleration loop can be implemented, so that the control system contains an

Sensors 2016, 16, 440; doi:10.3390/s16040440 www.mdpi.com/journal/sensors

Sensors 2016, 16, 440 2 of 13

acceleration closed-loop, a velocity closed-loop and a position closed-loop. Considering the installation
position of the sensors is limited to the narrow spces of the reverse side of the mirror, linear MEMS
accelerometers with small size, low weight and low power consumption are used in the acceleration
closed-loop. A detailed description of the method about using two linear accelerometers for measuring
the angular acceleration of the FSM is mentioned in [5]. The linear MEMS accelerometers used in this
?
paper are Model 1221 units with a bandwidth above 400 Hz and noise density of 14 ug{ Hz [6].
The acceleration feedback control (AFC) is a kind of high-precision robust control. It was
proposed by Studenny and Belanger in 1984 [7] and its application in mechanical arm control
was reported in a paper in 1991 [8]. In 1994, Bram de Jager studied the application of AFC in
tracking control [9]. Application research on acceleration feedback performed in torque control and
direct-driven mechanical arm shows that AFC is a highly effective technique [10,11]. In the previous
research, the actuator was a rotary motor. In theory, the acceleration open-loop transfer function of
this system driven by the rotary motor characterizes a low-pass filter. However, the actuators used
in the FSM control system are voice coil motors, and the acceleration open loop transfer function
includes a quadratic differential. In recent years, some scholars have used accelerometers and the
focal plane array (FPA) to implement closed-loop control of the FSM [5,12]. Accelerometers have
poor low-frequency response ability, while there exists a large drift in the double integral data of
accelerometers, which should be corrected with the FPA, thus it is difficult to use only accelerometers in
a FSM control system. The position information obtained from double integration of the accelerometers
fused with FPA data was used to achieve position loop control [5]. Strictly speaking, it is a single loop
position control method. Tang combined a CCD and accelerometers to implement dual closed-loop
control with acceleration and position [12]. In order to avoid the saturation of double integration, the
acceleration controller was designed as a bandpass filter. Therefore, there is still a quadratic differential
effect within the low-frequency range, and the disturbance suppression of the FSM control system for
low-frequency vibration is insufficient.
In this paper, under the conditiona of a low velocity closed-loop bandwidth, a three-closed-loop
control model (MEMS accelerometer feedback, MEMS gyroscope feedback and CCD feedback) is
proposed to replace the two-closed-loop control model (FOG feedback and CCD feedback) in the FSM
control system. A lag controller is used to accomplish the acceleration closed-loop control. The velocity
controller can be designed as a PI type controller, because the velocity open loop response with
AFC exhibits mainly an integrator. This paper is organized as follows: Section 2 presents a detailed
introduction to the control model of the FSM, mainly describing the CCD-based control structure and
the implementation of AFC. Section 3 discusses and analyzes the performance of MEMS accelerometers
and gyroscopes. Section 4 sets up the experiments used to verify this method. Concluding remarks are
presented in Section 5.

2. Control System Model

The configuration of the FSM control system is illustrated in Figure 1. The sensors include MEMS
accelerometers, fiber-optic or MEMS gyroscopes and a CCD. The controller is used to implement the
control algorithm. The driver actuates the voice coil motors to achieve the stabilization control of FSM.
The light source is used to simulate the target for the CCD.
 G p (s)   
U (s) s2 2 T s 1 (1)
2
 s 1 e
n n

The open loop natural frequency of FSMs, n is approximately between several Hz to tens of
Hz, and the damping factor  is much smaller than 1 [13].
Sensors 2016, 16, 440 3 of 13

Sensors 2016, 16, 440 3 of 13

( s ) K 1
G p (s)   2 
U (s) s 2 T s 1 (1)
2
 s 1 e
n n 

The open loop natural frequency of FSMs, n is approximately between several Hz to tens of
Hz, and the damping factor  is much smaller than 1 [13].

Figure 1.
Figure Configuration of
1. Configuration of the
the FSM
FSM control
control system.
system.

The  in Figure 2.
The traditional
mechanicalFSM part control system
of the FSM is a with theresonance
typical FOG is shown
element, while its electrical part including
the drivers and the voice coil motors of the FSM is a typical first-order inertial element, which is
1 
depicted as . Therefore, the FSM position open loop response can be expressed as follows [12]:
Te s ` 1 d ( s ) 
 (s )
Cvθpsq
( s) Gv (K
s) 1
U
G p psq “ “ ¨ (1)
Upsq s2 2ξ Te s ` 1
` s`1
$n 2 $n
Figure 1. Configuration of the FSM control system.
The open loop natural frequency of FSMs, $n is approximately between several Hz to tens of Hz,
and the damping factor ξ is much smaller than 1 [13].
The traditional FSM control system with the FOG is shown in Figure 2.
The traditional FSM control system with the FOG is shown in Figure 2.
Figure 2. FOG control system.

d ( s )  of Gp(s), which is depicted in

The FSM acceleration open loop response Gv(s) is a differential
Equation (1):
 (s )
Cv ( s ) Gv ( s )
 ( s ) K s
Gv ( s )   2 
U (s) s 2 T s 1 (2)
2
 s 1 e
n n
The velocity controller can be presented as follows:
Figure 2. FOG control system.
Figure 2. FOGcontrol system.
Cv  s   Gv
1
 s  Tlow  s  (3)
The FSM acceleration open loop response G Gvv(s) is a differential of G
differential of Gpp(s), which is depicted in
Tlow(s) is a low-pass filter to reduce vibrations and noise. The disturbance attenuation can be
Equation (1):
depicted as follows: .

Gv psq “   ( s )“ 2 KK
θpsq
¨
s
s (2)
Gv ( s) 
 s Upsq  2 ` 1 s`1  e `1
s 2ξ T s
U( s ) s2 $2n T s 1 (4)
(2)
 s 1  G$n s C
d   v  2 v  
 s sG 1 es FOG  
The velocity controller can be presented as  n
follows:
n

The velocity controller can be presented as follows:

Cv psq “ Gv´1 psq Tlow psq (3)
 
Cv s  G s Tlow s 1
v    
Tlow (s) is a low-pass filter to reduce vibrations and noise. The disturbance attenuation can be
(3)

depicted as is
Tlow(s) follows:
a low-pass filter to reduce vibrations and noise. The disturbance attenuation can be
.
depicted as follows: θ psq 1
. “ (4)
1 ` Gv psq Cv psq GFOG psq
  s 
θd psq
1
 (4)
 d  s  1  Gv  s  Cv  s  GFOG  s 
Sensors 2016, 16, 440 4 of 13
Sensors 2016, 16, 440 4 of 13

The second-order filter GFOG (s)represents

FOG(s) represents the
the characteristics
characteristics of
of the
the FOG
FOG [14]. The FSM control
system with the MEMS accelerometers and gyroscopes
gyroscopes is shown in Figure 3.
is shown in Figure 3.

d

Cv (s ) Ca (s ) Ga (s )

Figure
Figure 3.
3. MEMS
MEMS control
control system.
system.

The FSM acceleration open loop response Ga(s) is a quadratic differential of Gp(s), which is
The FSM acceleration open loop response Ga (s) is a quadratic differential of Gp (s), which is
depicted in Equation (1):
depicted in Equation (1):
..

( s )
θpsq KK s22
GGaa ( s ) Upsq “ s2 2 2ξ
psq “ ¨ (5)
U ( s ) s ` 2s ` 1 TTeess` 11 (5)
$2n2  $n s  1
n n
The ideal acceleration controller can be designed as the inverse transfer function Ga (s). In order
The ideal
to increase the acceleration
system gain,controller can be
an integrator designed
should as the The
be added. inverse
ideal transfer function
acceleration Ga(s). Incan
controller order
be
to increase the system
presented as follows: gain, an integrator should be added. The ideal acceleration controller can be
presented as follows: s 2 2ξ
` s`1
1 $n 22 $n Te s ` 1
Ca psq “ Ka ¨ ¨ s 2 ¨ (6)
s
2
K s  1 s2
1  ncancellation,
n Te s acceleration
1 (6)
Ca ( s )  K
Since the controller has complete pole-zero
a   the 2
closed-loop transfer
function may have high bandwidth theoretically. s K
However, theres are some disadvantages for the
controller, such as the saturated double integration, the worse disturbance
Since the controller has complete pole-zero cancellation, the acceleration suppression and stability
closed-loop transfer of
the system
function mayinduced
have by thebandwidth
high inaccurate control function
theoretically. which is there
However, derived arefrom
someinaccurate forGthe
fitting of
disadvantages a (s)
due to the noise and measuring error of accelerometers. To avoid these deficiencies,
controller, such as the saturated double integration, the worse disturbance suppression and stability the acceleration
controller
of C’a is
the system designed
induced by in
theEquation
inaccurate (7):control function which is derived from inaccurate fitting of
Ga(s) due to the noise and measuring error K of accelerometers. To avoid these deficiencies, the
a Te s ` 1
Ca1 “
acceleration controller C’a is designed in Equation ¨
(7): (7)
s T s`1 1

where Te s+1 is used to compensate phaseClag.' K a T es 1

a   integrator
The is used to compensate the quadratic
(7)
s constant
differential partly. The lag element with small time T1s  1 is used to filter the high-frequency noise.
The designed value of T1 should be smaller than 0.01, otherwise the bandwidth of the control system
where Tes+1 is used to compensate phase lag. The integrator is used to compensate the quadratic
will be too low. The closed-loop acceleration transfer function is expressed as follows:
differential partly. The lag element with small time constant is used to filter the high-frequency
noise. The designed value of T1 should Ka Ks be smaller than 0.01, otherwise the bandwidth of the control
system will be too low. The closed-loop ˆ 2
s acceleration
2ξ
˙ transfer function is expressed as follows:
pT1 s ` 1q 2
` s ` 1
$nK $n Ka Ks
Ga_closed “ a Ks “
K Ks
ˆ 2 ˙
1`  sa 2
2 ˙ s 1q
s 2ξ
s 1 ` Ka Ks
T1s  1  s2 2  2ξs  1
ˆ pT1 `
$Ks 2
`
$n
` (8)

pT1 s ` 1q  n2 ` n s `1  K a
n
Ga _ closed  $n $ 
K a Ks n KTa Ks  s2 2 
1“  1  2 
1s  ˆ s  1˙ K a Ks
 s 2
2  
1 1 s2 ` K aKn ` T 2ξ 
ˆ ˙
TT1s12 s13 ` 2 1 2 `s 2ξT 1`
n
s`1
(8)
$n 
 n n n $  $n $ n
K a Ks

T1 3  1 2T1  2  2 
s  2   s   K a K  T1   s 1
n2  n  n    n 
Sensors 2016, 16, 440 5 of 13

SensorsThis
2016,equation
16, 440 is similar to a band pass filter. Ga_closed is approaching zero at high frequency. 5 ofOn
13
the basis of analyzing of Equations (1) and (8), the denominator of the Equation (8) can be simplified
to KaKs + 1 at low frequency. And then Ga_closed can be simplified as follows:
This equation is similar to a band pass filter. Ga_closed is approaching zero at high frequency. On
Ga' analyzing
the basis of _ closed 
of Equations (1)Kand
a Ks (8), the denominator of theK
 a Ks
Equation K abe
(8) can Ks simplified
T1 3  1 And
to KaKs + 1 at low frequency. 2T1  2 G  2   2  K a Ks  1 (9)
s   then  K a K can
 s a_closed T1 be simplified
 s  1  K aas
K follows:
 T1   s 1
n2  n
2
n   n   n 
K Ks Ka Ks Ka Ks
1
Ga_closed “ ˙a ˆ « ˆKa, thus the value (9)
« KaKs is large
It is clear
T1that the acceleration controller has a high gain of
ˆ ˙ ˙
1 2ξT1 2ξ 2ξ K a Ks ` 1
s3 ` ` s2 ` Ka K ` T1 ` s`1 Ka K ` T1 ` s`1
enough to keep $n certain$ngain of
2 2 $n the closed loop at low
$n frequency which is $smaller
n than that of the
ideal closed loop. The analysis will be verified by the subsequent experiments. The acceleration
It is clear that the acceleration controller has a high gain Ka, thus the value of KaKs is large
disturbance attenuation is as follows:
enough to keep certain gain of the closed loop at low frequency which is smaller than that of the ideal

 by
closed loop. The analysis will be verified s  the subsequent
1 experiments. The acceleration disturbance
attenuation is as follows:   (10)
..d  s  1  Ga  s  Ca  s 
θ psq 1
.. “ (10)
The velocity open loop response 1 ` G
of the FSM with
θd psq a psq Ca psq
AFC at low frequency is depicted in
Equation
The (11):
velocity open loop response of the FSM with AFC at low frequency is depicted in
Equation (11): 1 K K K K a a
Gv _ a _ closed  Ga _ closed  
s1  s Ka K22
 K 1
KKa aKs (11)
Gv_a_closed “
s
Ga_closed “ T1s  1ˆ 2 2  2ξ s  1   K a Ks
˙ «
K Ks `1
(11)
pT1 s ` 1q   n `  n s ` 1  ` Ka Ks
s a
$n 2 $n
Therefore, the traditional PI controller can meet the velocity closed-loop control. The
Therefore,
disturbance the traditional
attenuation PI two
of the controller
loop can meet the
(velocity andvelocity closed-loop
acceleration loop)control.
control The disturbance
system can be
attenuation of the two
depicted as follows: loop (velocity and acceleration loop) control system can be depicted as follows:
.
θ psq  s  11 11
 ¨ (12)
.  “s 1 `1 GG 1 1 G s psq  sapsq
a  sC
 Ca apsq
s
a psq G Ca C (12)
θ psqd
d  s 1 a` aG psq
1 `CCv v spsq
s 1  Ga  sa Ca Csa psq

From Equation (12), we can see that the supersession

supersession bandwidth is subject to the acceleration
closed-loop bandwidth. Comparing Equations (4) and (12), the disturbance attenuation with MEMS
accelerometers and gyroscopes is better than that with FOG as long as the MEMS accelerometer
bandwidth is higher than that of FOG.
The multi-loop
multi-loop control
control structure
structure of
ofthe
theFSM
FSMwithwiththe
theMEMS
MEMSaccelerometers
accelerometers and
and gyroscope
gyroscope is
is shown
shown in Figure
in Figure 4. theFor
4. For FSMthe FSM system,
control control the
system, the isobjective
objective is to
to improve improve suppression
disturbance disturbance
suppression
performance.performance.

 d (s)

s2
 ( s )
d
( s)  (s)
C p (s) Cv ( s) Ca ( s ) Ga ( s )

Figure
Figure 4.
4. Stabilization
Stabilization control
control structure of multiple
structure of multiple closed
closed loops.
loops.
 Sensors 2016, 16, 440 6 of 13

Where Ga (s) is the acceleration open loop transfer function, Ca (s) is the acceleration controller,
.
Cv (s) is the velocity controller, Cp (s) is the position controller, θpsq is the angular velocity output, θd psq
..
is the disturbance angle and θd psq is the disturbance acceleration.
The disturbance suppression of the FSM with three closed loops can be expressed as follows:

1 Ga Ca
Cv
s 1 ` Ga Ca Ga Ca
« 1, «1 (13)
1 Ga Ca 1 ` Ga Ca
1` Cv
s 1 ` Ga Ca

θpsq 1
Eθ “ “
θd psq 1 1
1 ` Ga Ca ` Ga Ca Cv ` 2 Ga Ca Cv C p
s s
1 1 1
“ ¨ ¨
1 ` Ga Ca 1 Ga Ca 1 Ga Ca
1` Cv C
s 1 ` Ga Ca 1 s 1 ` Ga Ca v (14)
1` Cp
s 1 Ga Ca
1` Cv
s 1 ` Ga Ca
1 1 1
« ¨ ¨
1 ` Ga Ca 1 1
1 ` Cv 1 ` C p
s s
Therefore, the disturbance suppression of the system is equal to the product of the individual
suppression of three closed loops.
For the velocity open loop transfer function without AFC, the function sensitivity is equal to 1.
With AFC, the function sensitivity becomes:

pGa ` ∆Ga qCa Cv 1

ˆ ˙ ˆ ˙
Ga Ca Cv 1 Ga Ca Cv 1
¨ ´ ¨ { ¨
.
1 ` pGa ` ∆Ga qCa s 1 ` Ga Ca s 1 ` Ga Ca s Ga Ca 1
Sθ a
Ga “ « « (15)
∆Ga {Ga 2
p1 ` Ga Ca q 1 ` Ga Ca
.
Providing that the gain of the acceleration controller is large enough, Sθ a
Ga is far less than 1. When
the relevant structures and parameters of the system change greatly, the robustness of the velocity
closed-loop system with AFC will not be affected. The gain of the acceleration controller actually
exceeds 100, and so the system with the MEMS accelerometers and gyroscope feedback is robust.

3. Analyzing of the MEMS Accelerometers and Gyroscopes

The MEMS accelerometer and gyroscope have smaller size, lower weight, lower price and lower
power consumption than the FOG. Figure 5 and Table 1 depict the difference of the three types
of sensors.
Sensors 2016, 16, 440 7 of 13

Figure
Figure 5.
5. Picture of inertial
Picture of inertialsensors.
sensors.

Table 1. Comparison of the three types of sensors.

MEMS Gyroscope MEMS Accelerometer

FOG (FOG50)
(SCR1100-D02) (Model 1221)
Size Ø55 mm × 45 mm 25 mm × 20 mm × 8 mm 20 mm × 16 mm × 8 mm
Weight 250 g 25 g 20 g
Sensors 2016, 16, 440 7 of 13

Sensors 2016, 16, 440 7 of 13

Table 1. Comparison of the three types of sensors.

MEMS Gyroscope MEMS Accelerometer

FOG (FOG50)
(SCR1100-D02) (Model 1221)
Figure 5. Picture of inertial sensors.
Size Figure
Ø55 mm ˆ 45 mm5. Picture of inertial sensors.
25 mm ˆ 20 mm ˆ 8 mm 20 mm ˆ 16 mm ˆ 8 mm
Weight 250 g 25 g 20 g
Table 1. Comparison of the three types of sensors.
Price 50,000
Table Yuan
1. Comparison of the three1200 Yuan
types of sensors. 1900 Yuan
Power consumption 6W MEMS Gyroscope 0.15 W 0.1 W
MEMS Accelerometer
FOG (FOG50) MEMS Gyroscope MEMS Accelerometer
FOG (FOG50) (SCR1100-D02) (Model 1221)
(SCR1100-D02) (Model 1221)
TheSize
FOG-basedØ55 Inertial Measurement
mm × 45 mm Unit mounted on
25 mm × 20 mm × 8 mm the mirror
20 mmis× usually
base 16 mm ×used
8 mm as the
Size Ø55 mm × 45 mm 25 mm × 20 mm × 8 mm 20 mm × 16 mm × 8 mm
feedback component of the
Weight 250velocity
g closed-loop in25theg high precision FSM control
20 g system. The
Weight 250 g 25 g 20 g
bandwidth
Priceand noise of 50,000
the FOG have significant effects
Yuan on the performance of the
1200 Yuan FSM
1900 control system.
Yuan
Price 50,000 Yuan 1200 Yuan 1900 Yuan
The bandwidth
Power of the FOG for the latter experiment is about 500 Hz. The characteristics of the FOG
Power 6W 0.15 W 0.1 W
noise are depicted in Figure66.W
consumption 0.15 W 0.1 W
consumption
Fiber Gyroscope Noise Spectrum of Noise
Fiber Gyroscope Noise Spectrum of Noise
0.03 X: 100.4
0.1 0.03 Y:
X: 0.03103
100.4
0.1
Y: 0.03103

0.025
0.025
0.05
0.05
0.02
 /s /s

0.02
/s/s

Amplitude

0
Amplitude
Noise

0 0.015
Noise

0.015
X: 301.2
Y:
X: 0.009533
301.2
-0.05 0.01 Y: 0.009533
-0.05 0.01 X: 501.9
Y: 0.005917
X: 501.9
Y: 0.005917
0.005
0.005
-0.1
-0.1

1 2 3 4 5 6 7 0 500 1000 1500 2000 2500

1 2 3Time 4(s) 5 6 7 0 500 1000 1500
Frequency (Hz) 2000 2500
Time (s) Frequency (Hz)

Figure
Figure 6.
6. FOG
FOG noise
noise characteristics.
characteristics.
Figure 6. FOG noise characteristics.

The peak value of the FOG noise is equal to 0.25˝°/s, and the RMS value is about 0.037˝°/s. Aside
The peak value
value of
of the
the FOG
FOG noise
noiseisisequal
equaltoto0.25
0.25 °/s, the RMS
/s, and the RMS value
value isisabout
about0.037
0.037 °/s.
/s. Aside
from some “spikes”, the amplitude value of amplitude-frequency curve is smooth. The bandwidth
from some
some “spikes”,
“spikes”, the
theamplitude
amplitudevalue
valueofofamplitude-frequency
amplitude-frequencycurvecurveis is smooth.
smooth. TheThe bandwidth
bandwidth of
of the MEMS gyroscope for the latter experiment is less than 100 Hz, which is lower than that of the
of
thethe MEMS
MEMS gyroscope
gyroscope for for
the the latter
latter experiment
experiment is less
is less thanthan 100 which
100 Hz, Hz, which is lower
is lower than than that
that of theofFOG.
the
FOG. Figure 7 shows the characteristics of the MEMS gyroscope noise.
FOG.
FigureFigure 7 shows
7 shows the characteristics
the characteristics of the of the MEMS
MEMS gyroscope
gyroscope noise. noise.
Gyroscope Noise x 10 -3 Spectrum of Noise
-3
Gyroscope Noise x 10 Spectrum of Noise
0.05
0.05
2
2
Noise/s/s

 /s /s

1.5
Noise

1.5
Amplitude
Amplitude
Gyroscope

0
Gyroscope

0 1
1

0.5
0.5

-0.05
-0.05 2 4 6 0 500 1000 1500 2000 2500
2
Time4(s) 6 0 500 1000 1500 2000 2500
Frequency (Hz)
Time (s) Frequency (Hz)
Figure 7. MEMS gyroscope noise characteristics.
Figure
Figure 7.
7. MEMS
MEMS gyroscope
gyroscope noise
noise characteristics.
characteristics.

The peak value of the MEMS gyroscope noise is equal to 0.12 ˝ /s, and the RMS value is about
0.015 ˝ /s, which is smaller than that of the FOG. The MEMS gyroscope noise maily foucus on the
frequencies ranging from 0 to 200 Hz.
Sensors 2016, 16, 440 8 of 13

The peak value of the MEMS gyroscope noise is equal to 0.12 °/s, and the RMS value is about
Sensors 2016, 16, 440 8 of 13
0.015 °/s, which is smaller than that of the FOG. The MEMS gyroscope noise maily foucus on the
frequencies ranging from 0 to 200 Hz.
The practical bandwidth of the MEMS accelerometer
accelerometer is about 1000 Hz, which is higher than that
of the FOG. The characteristics
characteristics of
of the
the noise
noise are
are shown
shown in
in Figure
Figure 8.
8.

-3
x 10 Accelerometer Noise Spectrum of Noise
5 9

4
8
3
Accelerometer Noise /s2

7
2

Amplitude /s2
6
1
5
0
4
-1
3
-2
2
-3

-4 1

1 2 3 4 5 6 7 0 500 1000 1500 2000 2500

Time (s) Frequency (Hz)

Figure 8.
Figure 8. MEMS
MEMS accelerometer noise characteristics.
accelerometer noise characteristics.

The peak value of the MEMS accelerometer transformed noise is equal to 0.01 °/s2, and the RMS
The peak value of the 2MEMS accelerometer transformed noise is equal to 0.01 ˝ /s2 , and the RMS
value is about 0.00135 °/s . The amplitude value of amplitude-frequency curve is mainly smooth
value is about 0.00135 ˝ /s2 . The amplitude value of amplitude-frequency curve is mainly smooth aside
aside from some “spikes”.
from some “spikes”.
From the aforementioned analysis, the combination of MEMS gyroscopes and accelerometers
From the aforementioned analysis, the combination of MEMS gyroscopes and accelerometers
can be used to replace the FOG for accurately measuring the FSM’s high bandwidth disturbance,
can be used to replace the FOG for accurately measuring the FSM’s high bandwidth disturbance, and
and then the stabilized FSM with MEMS accelerometers and gyroscopes feedback can be
then the stabilized FSM with MEMS accelerometers and gyroscopes feedback can be accomplished
accomplished to replace the strap-down FSM for the line-of-sight (LOS) stabilization. Furthermore,
to replace the strap-down FSM for the line-of-sight (LOS) stabilization. Furthermore, the price of the
the price of the combination of two MEMS accelerometers and one MEMS gyroscope is 5000 Yuan,
combination of two MEMS accelerometers and one MEMS gyroscope is 5000 Yuan, which is one-tenth
which is one-tenth of the price of one FOG.
of the price of one FOG.
4. Experimental
4. ExperimentalVerification
Verification
The FSM
The FSM control
control system
system is is aa two-axis
two-axis system.
system. This
This experiment
experiment focuses
focuses on on one axis due
one axis due toto the
the
symmetry of the two axes. The experimental setup, which includes a disturbance platform, a stabilizeda
symmetry of the two axes. The experimental setup, which includes a disturbance platform,
stabilizedaplatform,
platform, laser light,a and
laseranlight,
image and an imagesystem
processing processing
(CCD), system
is shown(CCD), is shown
in Figure 9. Thein two
Figure 9. The
platforms
twodriven
are platforms are driven
by voice by voice
coil motors. coildisturbance
The motors. The disturbance
platform platform
simulates simulates theofdisturbance
the disturbance the carrier,
on which the fiber-optic gyroscope1 (FOG1) and the eddy current displacementdisplacement
of the carrier, on which the fiber-optic gyroscope1 (FOG1) and the eddy current sensor are used sensor
for
are used for disturbance measurement, FOG1 is used to measure disturbance
disturbance measurement, FOG1 is used to measure disturbance angular velocity of the platform, and angular velocity of the
platform,
the eddy isandusedthe eddy is used
to measure to measureangle
the disturbance the of
disturbance
the platform. angleTheofstabilized
the platform.
platformTheisstabilized
mounted
platform is mounted on the disturbance platform. The MEMS linear accelerometers
on the disturbance platform. The MEMS linear accelerometers mounted on the stabilized platform mounted on the
stabilized
are used toplatform
measure are the used
angular to measure
acceleration the of
angular acceleration
the platform. of the platform.
The fiber-optic gyroscopeThe2fiber-optic
(FOG2) is
used to measure the angular velocity of the stabilized platform. The light source is used toThe
gyroscope 2 (FOG2) is used to measure the angular velocity of the stabilized platform. light
simulate
source is used to simulate the target, and the mirror reflects the laser light into
the target, and the mirror reflects the laser light into the image processing system which detects the the image processing
system
LOS which
of the detectsplatform
stabilized the LOSand of provides
the stabilized platform
LOS error. and provides
The schematic LOS and
diagram error. The schematic
physical diagram
diagram and physical
are shown in Figure 9. diagram are shown in Figure 9.
The MEMS
The MEMS accelerometer
accelerometer bandwidth
bandwidth exceeds
exceeds 10001000 Hz,
Hz, and
and thethe angular
angular acceleration
acceleration cancan bebe
obtained from
obtained fromtwo two lineline accelerometers.
accelerometers. The bandwidth
The bandwidth of the of
FOG theexceeds
FOG 500 exceeds 500the
Hz, and Hz, and the
bandwidth
bandwidth of the MEMS gyroscope
of the MEMS gyroscope is less than 100 Hz. is less than 100 Hz.
From Equation
From Equation (5), (5), the
the open
open loop
loop acceleration
acceleration characteristic
characteristic of the FSM
of the FSM at at low
low frequency
frequency is is
similar to a second-order integrator. Figure 10 shows the open loop and the
similar to a second-order integrator. Figure 10 shows the open loop and the closed-loop acceleration closed-loop acceleration
response. The
response. The acceleration
acceleration closed-loop
closed-loop bandwidth
bandwidth exceeds
exceeds 700 700 HzHz and
and the
the closed-loop
closed-loop acceleration
acceleration
response includes
response includes aa little
littledifferential
differentialeffect
effectatatlow
lowfrequency
frequency which
whichis depicted
is depictedin Equation
in Equation (9).The
(9).
Sensors 2016, 16, 440 9 of 13

Sensors 2016, 16, 440 9 of 13

static acceleration noise of AFC is about 0.0028 °/s2, which is about twice than the noise of MEMS
Sensors 2016, 16, 440 9 of 13
accelerometers.
static acceleration
The static accelerationnoise
noiseof of
AFC is about
AFC 0.0028
is about 0.0028 ˝ /s2 , iswhich
°/s2, which about is
twice thantwice
about the noise
thanofthe
MEMS
noise of
MEMS accelerometers.
accelerometers.

Figure 9. Experimental apparatus: (a) global diagram; (b) local diagram.

Figure 9. Experimental apparatus: (a) global diagram; (b) local diagram.
Figure 9. Experimental apparatus:Bode
(a) Diagrams
global diagram; (b) local diagram.
10
Magnitude(dB)

0 Bode Diagrams
10
-10
Magnitude(dB)

0
-20 MEMS accelerometers opened
-10 -30 MEMS accelerometers closed
0 1 2 3
10 10 10 10
-20 MEMS accelerometers opened
200 MEMS accelerometers closed
-30
0 1 2 3
10100 10 10 10
Phase(deg)

0
200
-100
100
Phase(deg)

-200
0 1 2 3
0 10 10 10 10
Frequency(Hz)
-100
Figure 10. FSM acceleration response.
-200
0 1 2 3
10 10 10 10
Frequency(Hz)

Figure
Figure 10.
10. FSM
FSM acceleration
acceleration response.
response.
Sensors 2016, 16, 440 10 of 13
Sensors 2016, 16, 440 10 of 13

Theclosed-loop
The closed-loopbandwidth
bandwidthofofthe thesingle
singleFOG
FOGvelocity
velocity loop
loop is is about
about 220
220 HzHzin in Figure
Figure 11,11, which
which is
is higher than that of the MEMS gyroscope velocity loop with MEMS accelerometer
higher than that of the MEMS gyroscope velocity loop with MEMS accelerometer feedback. Therefore, feedback.
Therefore,
the the lower will
lower bandwidth bandwidth
producewill
someproduce
adversesome
effectadverse effectposition
on the outer on the outer
loop. position
The staticloop.
noiseThe
of
static noise of the single FOG velocity loop˝ is about 0.05 °/s, which is larger than
the single FOG velocity loop is about 0.05 /s, which is larger than that of the MEMS velocity that of the MEMS
loop
velocity
with AFC.loop
Thiswith AFC.
is due Thisdifferent
to the is due tobandwidth
the different bandwidth
of these of these sensors.
sensors.

Bode Diagrams

Magnitude(dB)
Magnitude(dB) 0

-20

-40
FOG feedback
MEMS sensors feedback
-60
0 1 2
10 0 10 1 10 2

200

100
Phase(deg)
Phase(deg)

-100

-200
0 1 2
10 0 10 1 10 2
Frequency(Hz)

Figure11.
Figure 11.FSM
FSMvelocity
velocityresponse.
response.

Figure 12 shows the CCD closed-loop response of the velocity closed-loop FSM with FOG
Figure 12 shows the CCD closed-loop response of the velocity closed-loop FSM with FOG feedback
feedback and with MEMS sensors feedback. The sample rate of the CCD is 100 Hz, which limits the
and with MEMS sensors feedback. The sample rate of the CCD is 100 Hz, which limits the bandwidth
bandwidth of the CCD closed-loop. The closed-loop bandwidths of two kinds of systems all reach
of the CCD closed-loop. The closed-loop bandwidths of two kinds of systems all reach about 16 Hz.
about 16 Hz. The static position error of the system with FOG feedback is approximately 0.113 pixel,
The static position error of the system with FOG feedback is approximately 0.113 pixel, while the static
while the static position error with MEMS sensors feedback is 0.115 pixel, which is a little larger than
position error with MEMS sensors feedback is 0.115 pixel, which is a little larger than that with FOG
that with FOG feedback. The primary causes are electronic and sensor noises.
feedback. The primary causes are electronic and sensor noises.

Bode Diagrams

0
Magnitude(dB)
Magnitude(dB)

-20

-40

-60 FOG feedback

MEMS sensors feedback
-80
0 1 2
10 0 10 1 10 2

200

100
Phase(deg)
Phase(deg)

-100

-200
0 1 2
10 0 10 1 10 2
Frequency(Hz)

Figure12.
Figure 12.CCD
CCDclosed-loop
closed-loopresponse.
response.

ItIt isis observed

observed from
from Figure
Figure 13
13 that
that the
the disturbance
disturbance suppression
suppression of
of the
the MEMS
MEMS accelerometer
accelerometer
closed-loop at low frequency is weaker than at medium frequency because
closed-loop at low frequency is weaker than at medium frequency because of the non-ideal of the non-ideal
acceleration controller C’’aa in Equation (5). Benefiting from the high closed-loop bandwidth, the
MEMS accelerometer loop keeps the disturbance suppression ability in the range of 200 Hz.
 Sensors2016,
Sensors 2016,16,
16,440
440 11ofof13
11 13

Bode Diagrams
acceleration controller C’ a in Equation
0 (5). Benefiting from the high closed-loop bandwidth, the

Magnitude(dB) Magnitude(dB)
Sensors 2016,
MEMS 16, 440
accelerometer loop keeps the disturbance suppression ability in the range of 200 Hz. 11 of 13
-10

-20
Bode Diagrams
-30
0
0 1 2 3
-1010 10 10 10

-20
50
-30

Phase(deg)
0 1 2 3
010 10 10 10

-50
50100 1 2 3
Phase(deg)

10 10 10
Frequency(Hz)
0

-50 suppression characteristics of the acceleration closed-loop.

Figure 13. Disturbance
0 1 2 3
10 10 10 10
Frequency(Hz)
The disturbance suppression characteristics of the FSM control system with the CCD and
gyroscope feedback are shown in Figure 14. The disturbance suppression with MEMS inertial
Figure 13. Disturbance
Figure 13. Disturbance suppression
suppressioncharacteristics
characteristicsofofthe
theacceleration
accelerationclosed-loop.
closed-loop.
sensors matches with that with FOG.
It is observed from Figure 14 and Table 2 that, with MEMS accelerometers feedback control, the
The disturbance
The disturbance suppression
suppression characteristics of of the FSM
FSM control system
system with the the CCD
CCD andand
disturbance suppression of the FSMcharacteristics
multi-loop controlthe system iscontrol
improved. Thewithexperimental result
gyroscopefeedback
gyroscope feedbackareare shown
shown in Figure
in Figure 14. disturbance
14. The The disturbance suppressionMEMSwith inertial
MEMS sensors
inertial
accords with the aforementioned analyzing about Equationssuppression
(4), (12) andwith
(14).
sensors matches with
matches with that with FOG.that with FOG.
It is observed from Figure 14 and Table 2 that, with MEMS accelerometers feedback control, the
disturbance suppression of the FSM multi-loopBode Diagrams
control system is improved. The experimental result
0
accords with the aforementioned analyzing about Equations (4), (12) and (14).
Magnitude(dB)Magnitude(dB)

-50 Bode Diagrams

0
FOG feedback
MEMS sensors feedback
-100
0 1 2
-5010 10 10

200 FOG feedback

MEMS sensors feedback
-100
100
Phase(deg)

0 1 2
10 10 10
0
200
-100
100
Phase(deg)

-200
0 1 2
010 10 10
Frequency(Hz)
-100

14. Disturbance
Figure 14.
Figure
-200 Disturbance suppression
suppressioncharacteristics.
characteristics.
0 1 2
10 10 10
Frequency(Hz)
It isTable 2. Comparison
observed between disturbance suppressions
from Figure 14 and Table 2 that, with MEMS of two kinds of FSM control
accelerometers systems.
feedback control, the
disturbance suppression
Frequencyof the FSM multi-loop
with FOG control system
with is improved.
MEMS The
Inertial experimental result
Sensors
Figure 14. Disturbance suppression characteristics.
accords with the aforementioned analyzing about Equations (4), (12) and (14).
1 Hz −81.39 dB −74.38 dB
TheTable
MEMS accelerationbetween
2. Comparison feedback loop hassuppressions
disturbance more than 700
of Hzkinds
two bandwidth
of FSM (shown in Figure 10).
10 Hz −46.39 dB −45.56 dB control systems.
However, the disturbance attenuation of the AFC is limited by the non-ideal acceleration feedback
20 Hz −35.77 dB −45.42 dB
controller belowFrequency
10 Hz. Therefore, thewith FOG
disturbance with MEMS
attenuation of theInertial
MEMSSensors
gyroscope feedback
40 Hz −25.57 dB −38.8 dB
system with AFC is1little
Hz weaker than −81.39
that ofdB
the FOG feedback system−74.38 below
dB 10 Hz and is better
100 Hz −21.97 dB −31.52 dB
between 10 to 100 Hz. 10 Hz −46.39 dB −45.56 dB
20 Hz −35.77 dB −45.42 dB
The MEMS acceleration feedback loop has more than 700 Hz bandwidth (shown in Figure 10).
40 Hz −25.57 dB −38.8 dB
However, the disturbance attenuation of the AFC is limited by the non-ideal acceleration feedback
100 Hz −21.97 dB −31.52 dB

The MEMS acceleration feedback loop has more than 700 Hz bandwidth (shown in Figure 10).
However, the disturbance attenuation of the AFC is limited by the non-ideal acceleration feedback
Sensors 2016, 16, 440 12 of 13

Table 2. Comparison between disturbance suppressions of two kinds of FSM control systems.

Frequency with FOG with MEMS Inertial Sensors

1 Hz ´81.39 dB ´74.38 dB
10 Hz ´46.39 dB ´45.56 dB
20 Hz ´35.77 dB ´45.42 dB
40 Hz ´25.57 dB ´38.8 dB
100 Hz ´21.97 dB ´31.52 dB

5. Conclusions
MEMS gyroscopes and accelerometers were introduced to replace the FOGs in the FSM control
system. The modeling of the FSM acceleration via linear accelerometers was discussed from the
viewpoint of its practical implementation. The simplification for implementing the AFC into the
FSM control system was presented mainly in terms of the closed-loop stability and error attenuation.
The algebraic expression shows that the AFC can effectively enhance the robustness of the closed-loop
control system. Three types of inertial sensors was discussed in terms of size, weight, price, power
consumption and performance. The experimental results showed that the MEMS accelerometers
feedback can effectively enhance the stabilization performance of the closed-loop control system, and
the combination of MEMS accelerometers and gyroscopes can act the role of the FOGs with lower cost
in the FSM control system.
Future work will concentrate on reducing the cost of the FSM control system under conditions of
good closed-loop performance. The use of only accelerometers may be an effective method, which will
be our next work. The accelerometer signal may be processed by two-dimensional digital filters [15].

Acknowledgments: We would like to gratefully acknowledge Jiang Bian for his optical technical support.
Author Contributions: Jing Tian is the head of the research group that conducted this study. He proposed
the original idea, analysed the theory, carried out the experiment, and writing this paper. Wenshu Yang and
Zhenming Peng contributed to the research through their general guidance and advice. Tao Tang joined analyzing
the theory and writing this paper. Zhijun Li joined the experiment.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Lawrence, M.G. Microradian-Level Inertial Line-of-Sight Stabilization Systems with Wide-Angle-Search
Capability. In Proceedings of the SPIE 1696, Controls for Optical Systems, Orlando, FL, USA, 21–22 April 1992.
2. Dickson, W.C.; Yee, T.K.; Coward, J.F.; McClaren, A.; Pechner, D.A. Compact fiber optic gyroscopes for
platform stabilization. In Proceedings of the SPIE 8876, Nanophotonics and Macrophotonics for Space
Environments VII, 88760L, San Diego, CA, USA, 21–25 August 2013.
3. Hilkert, J.M. A comparison of inertial line-of-sight stabilization techniques using mirrors. In Proceedings of
the SPIE 5430, Acquisition, Tracking, and Pointing XVIII, Orlando, FL, USA, 12 April 2004.
4. SCR1100-D02 Single Axis Gyroscope with Digital Spi Interface. Available online: http://www.murata.
com/~/media/webrenewal/products/sensor/gyro/scr1100/scr1100-d02%20datasheet%20v2%201.ashx?
la=zh-cn (accessed on 30 December 2015).
5. Ortiz, G.G.; Portillo, A.; Lee, S.; Ceniceros, J.M. Functional demonstration of accelerometer-assisted beacon
tracking. In Proceedings of the SPIE 4272, Free-Space Laser Communication Technologies XIII, San Jose, CA,
USA, 20 January 2001.
6. Model 1221 Low Noise Analog Accelerometer. Available online: http://www.silicondesigns.com (accessed
on 2 December 2012).
7. Studenny, J.; Belanger, P.R. Robot Manipulator Control by Acceleration Feedback. In Proceedings of the 23rd
IEEE Conference on Decision and Control, Las Vegas, NV, USA, 12–14 December 1984.
8. Studenny, J.; Belanger, P.; Daneshmend, L.K. A Digital Implementation of The Acceleration Feedback Control
Law on a PUMA 560 Manipulator. In Proceedings of the 30th IEEE Conference on Decision and Control,
Brighton, UK, 11–13 December 1991; Volume 3, pp. 2639–2648.
Sensors 2016, 16, 440 13 of 13

9. De Jager, B. Acceleration assisted tracking control. IEEE J. Control Syst. 1994, 14, 20–27. [CrossRef]
10. Han, J.; Tan, D.; Jiang, X. Joint Acceleration Feedback Control for Direct-drive Robot Decoupling.
Acta Autom. Sin. 2000, 26, 289–295.
11. Han, J.; He, Y.; Zhao, X. Mobile Robot System: Modeling, Estimation and Control; Science Press: Beijing,
China, 2001.
12. Tang, T.; Huang, Y.; Fu, C.; Liu, S. Acceleration Feedback of a CCD-Based Tracking Loop for Fast Steering
Mirror. Opt. Eng. 2009, 48. [CrossRef]
13. Cochran, R.W.; Vassar, R.H. Fast Steering Mirrors in Optical Control Systems. In Proceedings of the SPIE
1303, Advances in Optical Structure Systems, Orlando, FL, USA, 16–19 April 1990.
14. Lam, Q.M.; Stamatakos, N.; Woodruff, C.; Ashton, S. Gyro Modeling and Estimation of Its Random Noise
Sources. In Proceedings of the AIAA Guidance, Navigation, and Control Conference and Exhibit, Austin,
TX, USA, 11–14 Auguest 2003.
15. Ding, D.-W.; Du, X.; Li, X. Finite-Frequency Model Reduction of Two-Dimensional Digital Filters. IEEE Trans.
Autom. Control 2015, 60, 1624–1629. [CrossRef]

© 2016 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).