-
REMOTESENSINGSCANNINGMIRRORCONTROLLERDESIGN
M.Tech.Dissertation
Submitted in partial fulfillment of the requirements for the
degree of
MasterofTechnology
by
K Sunil Kumar
(09311023)
Under the Guidance of Prof. Hari B Hablani
Department of Aerospace Engineering
INDIAN INSTITUTE OF TECHNOLOGY, BOMBAY
June 2011
-
iii
Acknowledgement
I would like to express my sincere gratitude towards my guide
Prof. Hari B Hablani invaluable
guidance, motivation, constant support and encouraging
independent thinking. My thanks to all
my family members, who have always encouraged and supported me
in my pursuits. My sincere
thanks to all the people, who directly or indirectly have helped
in my project. My special thanks
to Khadar Basha.
Sunil Kumar K
Roll no: 09311023
-
iv
Abstract Remote sensing satellites play a major role in
monitoring the Earths air, ocean, and the land
from space. Satellite remote sensing is widely used as a tool in
many parts of the world for the
management of the resources. Many countries are concerned about
ecosystem and natural
resources, and to address this concern, series of remote sensing
satellites have been launched. A
thorough literature review of opto-mechanical sensors is
conducted which reveals that scanning
thematic mappers are used commonly for remote sensing of the
Earth. This project report
therefore is concerned with the control of scan motion of a
thematic mapper. Cross-track and in-
track profiles are formulated to scan successive rectangular
strips of the ground. MATLAB codes
are developed to generate these cross-track and in-track command
profiles of the sensor. To
impart this desired scan motion to the thematic mapper, several
controllers are designed,
satisfying the specification of turnaround and settling down the
instrument on a coasting rate in
11 ms with a tracking accuracy better than 1 millidegree. A
proportional-derivative (PD), a PD
with feed-forward, a proportional-integral-derivative (PID), a
PID with rate estimate, and a
double-lag compensator are designed to meet the performance
specifications. The time-domain
and the frequency-domain results are furnished in the report.
Sensor noise is modeled as white
noise with a standard deviation of 4% of the in-track angle
amplitude of 0.016 degree, that is, a
white noise with 0.6 millidegree (1 ). In order to achieve the
performance specifications, the payload scan angle is measured with
a frequency of 5 kHz in cross-track and 10 kHz in in-track.
The optical sensors resolvers and Hall-Effect sensors are
identified to provide these high-
frequency measurements. MATLAB SIMULINK software is used to
simulate these controllers.
An attempt is made to formulate multi-body dynamics of the
spacecraft bus and the hinged
sensor modeled as two articulated rigid bodies at an
articulation point.
-
v
Contents
List of figures vii
List of Tables xi
1 Introduction 1
1.1 Background and Motivation 1
1.2 Objective 2
1.3 Chapter wise content 2
2 Remote Sensing 4
2.1 Introduction 4
2.2 Applications of Remote Sensing 5
2.3 Remote Sensors 7
2.3.1 Classification of Remote Sensing 7
2.3.2 Opto-Mechanical Sensors 8
2.4 Literature Review on Thematic Mapper 11
2.4.1 LANDSAT program 11
2.4.2 Scan Mirror Assembly (SMA) 14
2.4.3 Scan Line Corrector (SLC) 17
3 Analysis of scan motion 20
3.1 Spherical Geometry method 20
3.2 Cross-track and In-track pointing commands for scanning
25
3.2.1 Extremities of In-track and Cross-track motions 31
3.3 Results 34
3.3.1 In-track and Cross-track results 34
3.3.2 Analytical results based on Spherical Geometry 36
-
vi
4 Scanning Mirror Controller Design Analysis 42
4.1 Introduction 42
4.2 Time Domain Design Analysis 43
4.2.1 PD controller Design 43
4.2.2 PID controller Design 58
4.2.3 Sample Rate selection 70
4.2.4 PID Rate Estimate controller 76
4.2.5 White Noise 82
4.3 Frequency Domain Analysis 85
4.3.1 Double Lag Compensator 85
4.4 Sensors and Actuators 95
4.5 Scanner ground pattern of different controllers 97
5 Multi-body Dynamics 101
6 Conclusion and Future Work 105
6.1 Conclusion 105
6.2 Future work 107
References 108
-
vii
List of Figures Figure 2.1: Different processes involved in
remote sensing 4
Figure 2.2: Schematic of operation of an opto-mechanical sensor
10
Figure 2.3.1 Landsat orbit 11
Figure 2.4: Landsat 7 satellite as viewed from sun side 12
Figure 2.5: optical systems and detector projection on ground
track 13
Figure 2.6: Scan Assembly Mirror 15
Figure 2.6: Scan Mirror Assembly operational requirements 16
Figure 2.7: Scan Mirror Assembly subassemblies 17
Figure 2.8: Scan line corrector function 18
Figure 3.1: Definition of angular relationships between
satellite, target and Earth center 20
Figure 3.2: Ground track and sensor scan motion on the surface
of the Earth 22
Fig 3.3: Relation between different unit vector 23
Figure 3.4: Relation between sub satellite coordinates and spot
coordinates
of sensor (mirror) 24
Figure 3.5: Relation between geometry as viewed from spacecraft
and from
the center of the Earth 25
Figure 3.6: Relation between geometry of satellite motion and
scan motion 29
Figure 3.7: Correction for satellite motion 32
Figure 3.8: Cross-track command 34
Figure 3.9: In-track command 35
Figure 3.10: Compensated motion with respect to LVLH frame
35
Figure 3.11: Scan pattern of the scanning mirror on the surface
of the Earth 36
Figure 3.12: Un-compensated scan motion of the scanning mirror
37
Figure 3.13: Compensated scan motion of the scanning mirror
38
Figure 3.14: Relation between true anomaly Vs spacecraft and
sensor
latitude and longitude 38
Figure 3.15 Variation of spot distance with respect to time
39
Figure 3.16 Variation of tilt and elevation angle with respect
to distance 39
Figure 3.17 Variation of elevation with respect to tilt angle
40
Figure 3.18 Variation of true anamoly with respect to cross
track 40
-
viii
Figure 3.19 Variation of Sensor Latitude and longitude with
respect
to space craft latitude 41
Figure 4.1: An open-loop control system 42
Figure 4.2: A system with feedback and feedforward control
43
Figure 4.3: A closed-loop system with unit feedback PD
controller in forward path 44
Figure 4.4: PD controller open loop frequency response plots
46
Figure 4.5: PD controller closed loop frequency response plots
47
Figure 4.6: PD controller error plots for cross-track motion
(0.9-damping ratio) 48
Figure 4.7: PD controller error plots for in-track motion at 0.9
damping ratio 50
Figure 4.8: Scan motion of the mirror using PD controller for
0.9 damping ratio
and enlarged scan motion 51
Figure 4.9: Simulink model of PD controller 52
Figure 4.10: In-track and cross-track loop with null
steady-state error for position,
Velocity, and parabolic input commands 53
Figure 4.11: Simulink model of feedforward PD controller 54
Figure 4.12: Cross-track and in-track error plots for PD
feedforward controller 55
Figure 4.13: Feedforward PD controller scanning motion output of
the sensor
(Damping ratio -0.9) 56
Figure 4.14: Scan motion of the sensor using PD feedforward
controller by
varying inertia 57
Figure 4.15: PID controller 58
Figure 4.16: Scan motion of the mirror using PID controller
(] = 0.9 and differentG ) and enlarged plot 62
Figure 4.17: cross-track error using PID controller (damping
ratio 0.9 and differentG ) and enlarged plot 63
Figure 4.18: In-track error using PID controller (damping ratio
0.9 and different G ) and enlarged plot 64
Figure 4.19: PID controller open-loop frequency response plots
65
Figure 4.20: PID controller closed-loop frequency response plots
66
Figure 4.21: Scan motion of the sensor using PID controller
(] =0.7 and different G values) and enlarged plot 68
-
ix
Figure 4.22: cross-track error (a) and in-track error (b) using
PID controller
(] =0.7 and different G values) 69
Figure 4.23: PID controller (zeta=0.9) with sample rate output
72
Figure 4.24: Cross-track error (a), In-track error (b) of PID
controller (zeta=0.9)
with sample rate 73
Figure 4.25: PD (zeta=0.9) feedback and feedforward with sample
rate 74
Figure 4.26: Cross-track error (a), intrack error (b) of PD
controller with sample rate 75
Figure 4.27: Bode diagram (a) and block diagram (b) for PID
compensator 76
Figure 4.28: PID rate estimate controller open-loop frequency
response 78
Figure 4.29: PID rate estimate controller closed-loop frequency
response 79
Figure 4.30: PID rate estimate controller (G =0.9 and ] =0.9)
Scan motion 80
Figure 4.31: In-track and cross-track error plot for PID rate
estimate controller 81
Figure 4.32: White noise with standard deviation-4% of the
in-track angle 82
Figure 4.33: PID (G =0.9 and ] =0.9) rate estimate with white
noise 83
Figure 4.34: cross-track and in-track error of PID (G =0.9 and ]
=0.9) controller
with white noise 83
Figure 4.35: PD (damping ratio-0.9) feedforward with white noise
84
Figure 4.36: Closed loop block diagram of the plant and second
order
lag compensator 85
Figure 4.37: Mirror scan motion using double lag controller and
enlarged plot 88
Figure 4.38: Cross-track and in-track error using double lag
compensator 89
Figure 4.39: Double lag compensator open loop frequency response
90
Figure 4.40: Closed loop frequency response of doable lag
compensator
for (a) cross-track (b) In-track 91
Figure 4.41: Double lag compensator nyquist plots 92
Figure 4.42: Scan motion with sensor noise of double lag
compensator
and enlarged plot 93
Figure 4.43: Cross-track and in-track error profiles with sensor
noise (White noise)
of double lag compensator 94
Figure 4.44: Controller output of scan motion on ground track
97
Figure 4.45: Variation of Elevation and Tilt angle Vs slant
range of
-
x
PD and PID controller 98
Figure 4.46: Elevation angle variation with respect to tilt
angle of
PD and PID controller 98
Figure 4.47: Variation of slant range with respect to time
99
Figure 4.41: Variation of Latitude and longitude of sensor with
respect to
latitude of spacecraft 99
Figure 4.42: Latitude and longitude of spacecraft and sensor
with respect
to true anomaly 100
Figure 5.1: System of two connected rigid bodies 101
-
xi
List of Tables 2.1: Landsat orbit parameters 12
2.2: SMA (Scan Mirror Assembly) characteristics 16
2.3: Scan line corrector (SLC) design parameters 18
4.1: Variation of 'DK ,'K P as damping coefficient changes for
cross-track motion 45
4.2: Variation of 'DK ,'K P as damping coefficient changes for
in-track motion 49
4.3: proportional, integral and derivative gain values for
cross-track controller 60
4.4: proportional, integral and derivative gain values for
in-track controller 61
4.5: Sample rate for different parameters of PID controller
71
4.6: q values for parameters damping ratio and integral gain
ratio 77
4.7: Raster scan parameters 86
6.1: comparison of different controller parameters 106
-
1
Chapter-1
Introduction
1.1 Background and Motivation:
A careful assessment of changes that occur in the environment
and in coastal ecosystems forms a
major milestone for effective coastal ecosystem management and
leads to sustainable utilization
of coastal resources. All these can be achieved only through the
collection of accurate, reliable
and comprehensive set of scientific data. Remote sensing is the
technique of deriving
information about objects on the surface of the earth without
physically coming into contact with
them. Remote sensing technology in recent years has proved to be
of great importance in
acquiring data for effective resources management and hence
could also be applied to coastal
environment monitoring and management.
The LANDSAT program provides repetitive acquisition of high
resolution multispectral data of
the Earth's surface on a global basis. The data from the Landsat
spacecraft constitute the longest
record of the Earth's continental surfaces as seen from space.
It is a record unmatched in quality,
detail, coverage, and value. The INSAT series was designed to
provide combined
telecommunications, direct TV broadcast, and meteorological
service, using INSAT services
early warnings of impending disasters (floods, storms) and could
directly reach to the civilian
population even in remote areas [26].
The Thematic mapper is a second generation Earth resources
sensor and it was placed in orbit
aboard LandsatD spacecraft in July of 1982. It is predecessors
of MSS (Multi Spectral scanner)
and the major difference is MSS collects data during west to
east direction along a scan
line.Thematic Mapper(TM) acquires data during both forward(west
to east) and reverse (east to
west)sweeps of its scanning mirror . After the TM, ETM+
(Enhanced Thematic Mapper plus) is
used and carries Scan Line Corrector (SLC) which is used to
correct scanning motion of the
scanning mirror. With the exception of a brief test on September
7, 2003, the SLC has remained
off since that time. The Landsat Data Continuity Mission (LDCM)
is the future of the landsat
satellite where Operational Land Imager (OLI) is used as payload
by replacing ETM+. For
-
2
almost 40 years, Landsat satellites have collected data of the
Earths surface to support the global
change research and applications [20].
These mission either LANDSAT, METEOSAT, INSAT, SPOT are using
opto-mechanical
sensors where mechanical scanning is involved. The scanning
mechanism also has to be
considered as one of the basic element of imaging systems in the
optical sensors. These
mechanisms currently facing challenges like controller design,
multi-body dynamics effect of
that mechanism on the spacecraft attitude. The current work
addresses one such scanning
mechanism and ground track of such mechanisms by using spherical
geometry methodology,
design a controller to track the scanning mechanism and analysis
of multi-body dynamics of the
spacecraft and payload.
1.2 Objectives:
Keeping in view of all these the present study has been planned
with the following objectives.
To study different kind of sensors used in Remote sensing
satellite and scanning mechanism by an opto-mechanical sensor.
To analyze the sensor coordinates relative to the subsatellite
point of the spacecraft by using spherical geometry method
To analyze the cross-track and in-track pointing commands for
scanning To design a controller to track the scan pattern of the
scan mirror by using different
controllers and analyze the controller with white noise
To analyze multi-body dynamics of the spacecraft and payload by
considering 3 Degrees of Freedom(DOF) for spacecraft and 2 DOF for
payload (scan mirror)
1.3 Chapter wise contents:This report organized in 6 chapters.
First chapter presents introduction, background and
objectives. Second chapter gives a brief overview of remote
sensing followed by classification of
remote sensing and different kinds of remote sensors. Detail
description on opto-mechanical
sensors, thematic mapper and literature review on Landsat is
discussed in this chapter. The third
chapter presents the analysis of scan motion by using spherical
geometry method to find out
sensor position (sensor coordinates) on the Earth fixed frame.
The in-track and cross-track
-
3
analysis for scan motion is briefly discussed in this chapter
and implement these commands into
spherical geometry analysis to get sensor coordinates.
Fourth chapter describes about different types of controller
used to track the scan motion of the
mirror. The time domain and frequency domain design of
controllers followed by different
sensors used to feedback the position of the scanning mirror is
briefly discussed. Sample rate
selection for feedback the signal and White noise also included
with the sensor, to make more
realistic investigation on the controller behavior. Fifth
chapter discusses the multi-body
dynamics of the spacecraft with articulate payload with two
degrees of freedom and some future
work related to multi-body dynamics also specified in this
chapter. A summary of the results,
conclusions and a few recommendations for future work are
presented in Chapter 6.
-
2.1 Intr
Remote
earth wit
making
platforms
recording
or digital
When el
transmitt
mostly in
the photo
function
basic pro
distingui
surfaces
the worl
containin
roduction:
e sensing is
thout coming
observations
s (aircraft an
g the observ
l data on ma
ectromagnet
ted through t
n the form o
ographic film
of the surfac
operty whic
sh between
using remot
ld for the m
ng reefs, isla
Figure 2.
:
the techniq
g into physic
s using sen
nd satellites)
vations on a
gnetic tapes)
tic radiation
the surface,
of heat. It is
m or digital
ce is describ
ch allows id
different sp
te sensing. S
management
ands, mangro
1: Different
Ch
Remo
que of derivi
cally contact
nsors (camer
), which are
suitable me
) [4].
n falls upon
and the rest
that reflecte
l sensor. Th
bed as proce
dentification
pectral signa
Satellite remo
t of the res
oves, shoals
processes in
hapter-2
ote sensin
ing informat
t with the obj
ras, scanner
e at a consid
dium (image
a surface, s
t is reflected
ed and emitt
he intensity
ssing a char
n of an obj
atures, then
ote sensing i
sources and
and nutrient
nvolved in re
ng
tion about o
jects under s
rs, radiomet
derable heigh
es on photog
some of its
d. Surfaces a
ted radiation
and wavele
acteristic S
ect. If an i
it will be po
is widely us
activities w
t rich waters
emote sensin
objects on th
study. This
ter, radar e
ht from the e
graphic film
energy is ab
also naturall
n which is re
engths of the
pectral Sign
instrument c
ossible to m
sed as a tool
within the c
.
ng [27]
he surface o
process inv
tc.) mounte
earth surface
ms and video
bsorbed, som
ly emit radia
ecorded eith
e radiation
nature, this i
can identify
map the exte
in many par
continental
4
of the
olves
ed on
e and
otapes
me is
ation,
her on
are a
is the
y and
ent of
rts of
shelf
-
5
The first requirement (A in fig.2.1) for remote sensing is to
have an energy source which
illuminates or provides electromagnetic energy to the target of
interest. As the energy travels (B)
from its source to the target, it will come in contact with and
interact with the atmosphere it
passes through. This interaction may take place a second time as
the energy travels from the
target to the sensor. Once the energy makes its way to the
target through the atmosphere, it
interacts with the target depending on the properties of both
the target (C) and the radiation.
After the energy has been scattered by, or emitted from the
target, we require a sensor (remote -
not in contact with the target) (D) to collect and record the
electromagnetic radiation. The energy
recorded by the sensor has to be transmitted (E), often in
electronic form, to a receiving and
processing station where the data are processed into an image
(hardcopy and/or digital). The
processed image is interpreted (F), visually and/or digitally or
electronically, to extract
information about the target which was illuminated. The final
element of the remote sensing
process is achieved when we apply (G) the information we have
been able to extract from the
imagery about the target in order to better understand it,
reveal some new information, or assist
in solving a particular problem [27].
There are two main types of remote sensing: passive remote
sensing and active remote sensing.
In passive remote sensing, passive sensors are used and they
detect natural radiation that is
emitted or reflected by the object or surrounding area being
observed. Reflected sunlight is the
most common source of radiation measured by passive sensors.
Examples of passive remote
sensors include film photography, infrared, charge-coupled
devices, and radiometers.
Active collection, on the other hand, emits energy in order to
scan objects and areas whereupon a
sensor then detects and measures the radiation that is reflected
or backscattered from the target.
RADAR is an example of active remote sensing where the time
delay between emission and
return is measured, establishing the location, height, speeds
and direction of an object.
2.2 Applications of Remote Sensing: Coastal areas, the place
where the waters of the seas meet the land are indeed unique places
in
our global geography. They are endowed with a very wide range of
coastal ecosystems like
mangroves, coral reefs, lagoons, sea grass, salt marsh, estuary
etc. They are unique in a very real
economic sense as sites for port and harbor facilities that
capture the large monetary benefits
-
6
associated with waterborne commerce and are highly valued and
greatly attractive as sites for
resorts and as vacation destinations. The combination of
freshwater and salt water in coastal
estuaries creates some of the most productive and richest
habitats on earth; the resulting bounty
in fishes and other marine life can be of great value to coastal
nations. In many locations, the
coastal topography formed over the millennia provides
significant protection from hurricanes,
typhoons, and other ocean related disturbances. But these values
could diminish or even be lost,
if they are not managed. Pollution of coastal waters can greatly
reduce the production of fish, as
can degradation of coastal nursery grounds and other valuable
wetland habitats.
To achieve this, an understanding of the coastal processes that
influence the coastal
environments and the ways in which they interact is necessary.
It is advantageous to adopt a
holistic or systematic approach for solving the coastal
problems, since understanding the
processes and products of interaction in coastal environments is
very complicated. A careful
assessment of changes that occur in the coastal environments and
in coastal ecosystems forms a
major milestone for effective coastal ecosystem management and
leads to sustainable utilization
of coastal resources. All these can be achieved only through the
collection of accurate, reliable
and comprehensive set of scientific data. Remote sensing
technology in recent years has proved
to be of great importance in acquiring data for effective
resources management and hence could
also be applied to coastal environment monitoring and management
[5].
Agriculture plays major role in economies of both developed and
undeveloped countries. The
production of food is important to everyone and producing food
in cost effective manner is the
goal of every farmer, regional agricultural agency. A farmer
needs to be informed to be efficient
and that includes having the knowledge and information to forge
a viable strategy for forming
operation. Satellite and air born images are used as mapping
tools to classify crops examine their
health and viability, and monitor farming practices. Agriculture
applications of remote sensing
includes crop type assessment, crop type classification, crop
estimation, mapping of soil
characteristic and management practices.
Forests are valuable resources providing food, shelter, wild
life habitat, fuel, and daily supplies
like medicinal ingredients and paper. Forest plays a major role
in carbon dioxide supply and
exchange, acting as key link between atmosphere, geosphere and
hydrosphere. International and
domestic forestry applications where remote sensing can be
utilized include sustainable
-
7
development, biodiversity, monitoring deforestation, wild life
habitat assessment and other
environmental concerns.
Remote sensing is used as a tool to extract information about
land surface structure, composition.
Multispectral data can provide information on lithology, rock
composition based on spectral
reflectance. Hydrology is related to many other applications of
remote sensing, particularly
forestry, agriculture and land cover, since water is a vital
component in each of these disciplines.
Land use of application of remote sensing are natural resource
management, base line mapping
for GIS input, routing and logistics planning for seismic,
research, extraction activities and
identification of landing strips, roads, clearing, bridges,
land/water interface [27].
2.3 Remote Sensors: Remote sensors are the instruments that
measure the properties of electromagnetic radiation
leaving a medium due to scattering or emission. Information
could be collected over a special
extent including the angular dependence of the observation and
as a function of distance along
the line of sight of the instrument.
2.3.1 Classification of Remote Sensors:
As mentioned earlier that passive remote sensing used passive
sensors which sense natural
radiation, either emitted or reflected from the Earth. It is
convenient to classify sensors as those
operating in optical-infrared region and those operating in the
microwave region. Imaging
sensors give 2-dimensional spatial distribution of the emitted
or reflected intensity of
electromagnetic radiation, while non-imaging sensors measure the
intensity of radiation within
field of view. In active remote sensing, active sensors are used
to produce electromagnetic
radiation of a specific wavelength or band of wavelengths .The
interaction of this radiation with
the target could then be studied by sensing the scattered
radiation from the targets [5].
An optical-infrared (OIR) sensor covers a wavelength region
extending from 0.4m to 20m.
The reception and analysis are carried out by instruments like
lenses, mirrors, prisms. These
sensors are further classified into photographic and
electro-optical. In the photographic system,
the images are formed directly on to a film where as in
electro-optical sensors; the optical image
is first converted into an electrical signal and further
processed to record the data.
-
8
The microwave region of interest to remote sensing covers the
electromagnetic radiation of
wavelength. However these sensors can operate irrespective to
the radiation of the sun. Thus, the
microwave sensors can operate during day and night. The
atmosphere is more transparent to the
microwaves, than to optical rays, thus providing weather
monitoring capability. But the current
work is dealt with the optical-infrared sensors only.
In photographic cameras, where image is formed in conventional
manner by lens, recording are
restricted to particular wavelength within the visible region.
The camera may be an instrument in
which the image is captured in a film or a charge coupled device
(CCD). Alternately, it may be
like television camera which could be referred as return beam
vidion (RBV) cameras in which
image is converted into signal that is superimposed on a carrier
wave and transmitted to distant
receiver. In case of non-photographic cameras or electro-optical
sensors, no lens is involved that
means no image is formed or image is formed in completely
different manner from the method
used in a camera with lens and instrument is able to operate at
longer wavelengths in the infrared
part of the spectrum.
Multispectral scanner (MSSs) is optical-mechanical scanner that
is widely used in remote
sensing and it is able to operate both in the visible and
infrared ranges of wavelengths. MSS is
like a simple radiometer; first it splits the beam of received
radiation into a number of spectral
ranges and secondly by adding the important features of
scanning. The image is not formed all at
once as it is in camera but is built up by scanning. In most
cases, this scanning is achieved using
a rotating mirror; in other either whole satellite spin or a
push-broom technique using a one-
dimensional CCD array is employed. The push broom scanner is one
of the scanning systems
that have no moving parts and its one-dimensional array of CCDs
that is used in place of a
scanning mirror to achieve cross-track scanning.
2.3.2 Opto-Mechanical Scanners:
Most of the limitations associated with photographic and TV
imaging systems are overcome in
opto-mechanical scanners. Mechanical scanning is involved in
some of the Opto-mechanical
sensors like MSS, Thematic mapper (TM), Enhanced Thematic Mapper
plus (TM). Most of the
remote sensors acquire data using scanning systems, which employ
a sensor with a narrow field
of view that sweeps over the terrain to build up and produce a
two-dimensional image of the
-
9
surface. A scanning system used to collect data over a variety
of different wavelength and is
most commonly used in scanning system.
The spectral range of photographic systems is restricted to the
visible and near-infrared regions
while MSS systems can extend this range into the thermal
infrared. They are also capable of
much higher spectral resolution than photographic systems.
Multi-band or multispectral
photographic systems use separate lens systems to acquire each
spectral band. This may cause
problems in ensuring that the different bands are comparable
both spatially and radiometrically
and with registration of the multiple images.
The radiation emitted or reflected from the scene is intercepted
by a scan mirror, diverts the
radiation to a collecting telescope. In the normal case the scan
mirror is inclined at 45deg. to the
optical axis of the telescope. The telescope focuses the
radiation on to a detector. The detector
receives radiation from an area on the ground which is
determined by detector size and focal
length of the optics. By rotating the scan mirror, the detector
starts looking at adjacent picture
elements on the ground. There are two main methods of scanning
employed to acquire
multispectral image data across-track scanning and along-track
scanning [5].
The rotation of scan mirror collects radiation from a strip on
the ground whose width equals to a
pixel and is at right angles to the nadir track. If the rate of
rotation of the scan mirror is adjusted
such that by the time the platform moves through one picture
element, the scan mirror set to start
of the next scan line, then successive and contiguous scan lines
can be produced. Thus in the
cross-track direction, the information is collected from each
pixel to produce one line of image
and in the along-track direction, successive lines of image in a
contiguous fashion are produced
by the platform motion as shown in fig.2.2. To produce
multispectral imaginary, the energy
collected by the telescope is channeled to a spectral dispersing
system-spectrometer. Such
systems which can generate imagery simultaneously in more than
one spectral band are called
Multi-spectral Scanners (MSS).
In opto-mechanical imager scanning can be carried out either in
the object plane or in the image
plane. In the image plane scanner, the scan mirror is kept after
the collecting optics near to the
focal plane and the mirror detects each point in the focal plane
to the detector. Such a system
requires the collecting optics corrected for the total field of
view, which is quite difficult. Though
-
image pl
scanners,
not gener
In the ca
reflects t
direction
position.
scan effi
upwards,
increase
swath. L
series use
Figure 2
lane scannin
, due to large
rally used.
ase of object
the radiation
n of the rays
This kind o
ciency can b
, the second
the scan eff
LANDSAT s
ed VHRR (V
2.2: schemat
ng has bee
e field corre
t plane scan
n to the col
s at the coll
of scanning
be increased
d mirror can
ficiency, inst
series used o
Very High R
tic of operati
n used in
ction require
nning, the ra
llecting teles
lecting optic
has been ex
d by adding
n scan the g
tead of rotat
opto-mechan
Resolution Ra
ion of an opt
some of th
ed for the tot
ays from the
scope. Here
cs remains th
xtensively u
a second m
ground. In t
ting, the mir
nical sensors
adiometer) a
to-mechanic
he early opt
tal field of v
scene fall o
for the ray
he same irre
sed in the m
mirror so tha
the case of
rrors are osc
s like TM, E
as Remote se
cal sensor [5]
to-mechanic
view, image p
on to the sc
ys reaching
espective of
meteorologic
at while the
high resolu
cillated to co
ETM as sen
ensor.
]
cal multi-spe
plane scanni
an mirror, w
the detector
f the scan m
cal payloads
first mirror
ution imager
over the req
nsors and IN
10
ectral
ing is
which
r, the
mirror
. The
r sees
rs, to
quired
NSAT
-
11
2.4 Literature Review on Thematic Mapper:
The TM relies heavily on the technology of the MSS, it is
designed to achieve finer special
resolution, sharper spectral separation, and greater radiometric
accuracy and resolution. The
following section gives us detailed description of scanning
parameters of TM which was used in
LANDSAT. The TM acquires data in seven spectral bands covering
the visible, near-infrared,
middle-infrared, and thermal infrared regions of the
electromagnetic spectrum.
2.4.1 LANDSAT program:
The Landsat Program has provided high
spatial resolution data of the Earth's surface to
a broad and varied user community, including
agribusiness, global change researchers, state
and local governments, commercial users,
military, and the international community.
Landsat images provide information meeting
the broad and diverse needs of business,
science, education, government, and national
security.
The mission of the Landsat Program is to
provide repetitive acquisition of high
resolution multispectral data of the Earth's surface on a global
basis. Landsat represents the only
source of global, calibrated, high spatial resolution
measurements of the Earth's surface that can
be compared to previous data records. The data from the Landsat
spacecraft constitute the
longest record of the Earth's continental surfaces as seen from
space. It is a record unmatched in
quality, detail, coverage, and value.The Landsat platforms carry
multiple remote sensor systems
and data relay systems along with attitude-control and
orbit-adjust subsystems, power supply,
and receivers for ground station commands and transmitters to
send the data to ground receiving
stations.
-
12
The injected spacecraft is a 1940kg satellite designed for a
705-km, sun synchronous, earth mapping orbit with a 16-day
repeat cycle. Its payload is a single nadir-pointing
instrument,
the Enhanced Thematic Mapper. The LANDSAT-D is an-
experimental Earth-recourses monitoring system that utilize
the
capabilities of TM, which is the principal instrument on the
LANDSAT-D spacecraft. The LANDSAT-D spacecraft was
launched into a circular near-polar sun-synchronous orbit
[25].
Table 2.1: Landsat orbit parameters [25]
Figure 2.4: Landsat 7 satellite as viewed from sun side [25]
The TM design provides for a nadir-viewing, eight-band
multispectral scanning radiometer
capable of providing high-resolution image information of the
Earth's surface when operated
from Landsat-7, a 3 axis stabilized spacecraft located in a near
polar, sun-synchronous and
circular orbit at a 705 km nominal altitude, with an orbit
inclination of 98.2 degrees. The TM is
designed to collect, filter and detect radiation from the Earth
in a swath 185 km wide as it passes
overhead and provides the necessary cross-track scanning motion
while the spacecraft orbital
motion provides an along-track scan.
Altitude 705km
Inclination 98.2 deg.
Repeat period 16 days
Orbits/repeat
period
233
Trace spacing 172 km
-
13
The Landsat TM, MSS instrument achieves its 185-kilometer
cross-track ground swath by using
a bi-directional scan mirror to traverse the instrument
line-of-sight through a 15.4-degree cross-
track field of view. On descending passes, the mirror
alternately scans west-to-east (forward
scan) and then east-to-west (reverse scan). The active
Earth-imaging portion of each scan lasts
for, nominally, 60,743 micro seconds. During this time, the
Landsat spacecraft advances
approximately 410 meters down track. If uncompensated, this
spacecraft motion would add an
along-track component to the scan mirror's cross-track
trajectory, leading to a zig-zag scanning
pattern that overlaps where one scan end adjoins the next scan
start, and leaves gaps where one
scan end adjoins the previous scan start. The maximum gap at the
edge of the TM swath is the
distance traveled in two active scans plus one mirror turnaround
period (11.57 milliseconds), less
the width of one scan (480 meters). This is approximately 420
meters. The thematic mapper
acquires data in seven spectral bands covering visible, near
Infrared, middle infrared, thermal
infrared. [14].
The scanning mechanism operating at 7Hz
provides the cross-track scan; where as the
orbital velocity provides the scan along the
track.A Ritchey-Chretien telescope focuses
the energy onto a pair of motion
compensation mirrors (i.e. scan line
corrector) where it is redirected to the focal
planes as shown in the figure 2.5. The scan
line corrector is required due to the
compound effect of a long-track orbital
motion and cross-track scanning which leads
to significant overlap and under lap in
ground coverage between successive scans.
Figure 2.5: optical systems and detector projection on ground
track [14]
-
14
To achieve the required level of scan mirror angular position
knowledge and control, the TM
scan mirror assembly includes a device, known as the scan angle
monitor (SAM). The SAM
provides timing pulses to the scan mirror control electronics at
fixed angles representing the
beginning, middle, and end ofeach earth-view active scan [15].
The SAM timing pulses serve
two functions: 1) to initiate and halt the collection of
earth-view image data during the active
scan time and 2) to allow the scanmirror control electronics to
monitor the scan mirror velocity
and maintain the desired active scan time. When operating in SAM
mode, the SAM timing
pulses are used to compute the deviations from the ideal scan
start to midscan and midscan to
scanend times for each active scan so that the active scan times
arenearly the same for each
mirror sweep. In this operating mode, the active scan angle is
held fixed by the SAM and the
amount oftime required to traverse this fixed angle is
controlled to withina few detector sample
time periods. The amount of time the scanmirror takes to reverse
direction between scans is not
controlled [22].
2.4.2 Scan Mirror Assembly (SMA): The heart of the TM is the
Scan Mirror Assembly (SMA). The SMA performs the following four
primary functions:
Delivering a linear earth scan into the TM telescope.
Controlling the active scan time. Synchronizing the video data at
the multiplexer. Collecting scan error data for correction and
engineering data for analysis on the ground.
An object-space scan mirror sweeps the TM line of sight back and
forth seven times each second
in a direction normal to the orbital ground track to form a
raster of 16 lines in bands1 through 5
and band7, and four lines in band6. Data are collected during
both the forward (west to east) and
reverse scan (east to west) scans with a scanning efficiency of
85percent. The scan velocity is
essentially constant during both directions of scan for two
reasons:
1) Torque is applied to the scan mirror only during the
turnaround times (not during the active
scans)
2) The scan mirror floats in a nearly inertia-free manner during
the active scan through the use
of magnets through the use of magnets which compensate for the
flex pivot spring forces.
The bands are displaced with respect to each other in their
focal planes and in effect are scanned
across the same earth point at different times. Because of this
characteristic during active scan,
-
15
linearity and stability of the scan rate important. While
hitting turnaround bumper springs, the
Scan Mirror Assembly (SMA), the mechanical scanning mirror,
produces as much as 23,000 gs
peak and impart 4 ft-lb of torque to the spacecraft; less than
11 milliseconds later, the scanning
mirror must start collecting picture image data not influenced
by structural self-produced
vibration [15].
The Scan Mirror Assembly (SMA) provides the cross-track scanning
motion to develop the 185-
km long scene swath. The SMA consists of a flat mirror supported
by flex pivots on each side
(which have compensators to equalize pivot reaction torque), a
torque (induction motor), and a
scan angle monitor (SAM), 2 leaf spring bumpers and scan mirror
electronics (SME). The
motion of the mirror in each direction is stopped by the bumper,
and is boosted by precision
torque pulses during the turnaround period. The amount of torque
applied is controlled by the
SME microprocessor as determined from the SAM mirror angle
pulses.
Figure 2.6: Scan Assembly Mirror [15]
The resulting active scan in each direction is closely
controlled to 60,743 micro seconds. SAM
mirror angle pulses are used by the multiplexer to synchronize
the detected scene data. There are
redundant sets of scan mirror electronics (SME), SME 1 and SME
2. SME 2 is an identical back
up electronic package to SME-1. Additionally, both SMEs have a
primary SAM mode of
operation and a back up Bumper Mode of operation.
-
16
Figure 2.6: Scan Mirror Assembly operational requirements
[15]
Activescanamplitude
7.695degrees
Scanperiod
142,925microseconds
Scanningfrequency
6.997Hz
Activescanperiod
60,743microseconds
Turnaroundtime
10,719microseconds
Objectplanescanrate
4.42191rad/sec
Table 2.2: SMA characteristics
The scan mirror oscillates by rebounding between the bumper
assemblies. If there are no
restoring torquess between rebounds, a constant angular velocity
will be obtained. The restoring
torque due to the spring constant and nonlinearity of the
flexural pivot is compensated almost
-
perfectly
impulses
The desi
readings
proportio
scan erro
The SMA
portion o
(SAM) s
scan line
relative t
2.4.3: S
A second
detector
scan mir
previous
structura
detectors
(SLC) is
y by an E-ma
are supplied
ired duration
weighted b
onal gain, di
or are teleme
A was desig
of each scan
tart, mid-sca
e. The signa
to the SMA/T
Scan Line
d line scann
swath cause
rror turnarou
set. The S
l influence.
s and bands,
an electro-o
agnet set in t
d electrically
Figure 2.
n of the torq
by the appr
irect proport
etered to the
gned so tha
line) has lin
an and SMA
als are produ
TM frame.
Corrector
ning mirror
ed by space
und so that
LC, operati
During the
thereby pro
optical mech
the TM moun
y during each
7: Scan Mir
que pulses i
ropriate gain
tional gain.
ground for s
at, ideally, n
ne length du
A end pulses
uced at the
r (SLC):
, called the
ecraft orbital
the next se
ng at 14 H
turnaround,
viding the in
hanism comp
nt. To maint
h turnaround
rror Assembl
is computed
ns. The thre
The comput
scan nonline
no external
uration of 60
s appear at t
instant the
Scan Line
l motion. Th
et of scan li
Hz, must als
, the main c
nternal detec
posed of two
tain the ener
d.
ly subassemb
d by the mic
ee gains are
ted first-half
earity correct
nonsynchron
0743 micros
the beginnin
SAM passe
Corrector
he SLC jum
nes is conti
o transpose
calibration s
ctor calibrati
o parallel nic
rgy level of t
blies [15]
croprocessor
e direct inte
f scan error
tion.
nous disturb
econds. Sca
ng, mid-scan
es through a
(SLC), com
mps around a
iguous and
optical ima
shutter swee
ion. The Sca
ckel-plated b
the mirror, to
r from scan
egral gain,
and second
bance (the a
an Angle Mo
n and end of
a particular a
mpensates fo
ahead durin
parallel with
ages withou
eps across al
an Line Corr
beryllium mi
17
orque
time
cross
half-
active
onitor
f each
angle
or the
g the
h the
ut the
ll the
rector
irrors
-
18
set at an angle on a shaft. The shaft, driven by a torquer which
has primary and redundant
tachometers, rotates about an axis normal to the axis of the
scan mirror in a saw tooth fashion.
The SLC is positioned behind the primary optics and compensates
for the in-track motion of the
spacecraft that occurs during an active SMA cross track scan. A
rectilinear scan pattern is
produced using the SLC instead of the zig-zag pattern that would
be produced without it fig.2.8.
Scan frequency 13.99Hz
Scan rate in object space 9,610 rad/sec
SLC rotation rate 576.7 rad/sec
SLC linear scan angle 32.02 rad
Table 2.3: Scan line corrector (SLC) design parameters
Figure 2.8: Scan line corrector function [14]
-
19
The SLC retraces, moving from its end-of-scan full aft position
to its scan-start full forward
position, during the turnaround time while the scan mirror is
changing directions. The SLC
adjusts the TM along-track pointing so that, if the scan mirror
were not moving, it would view a
constant location during each active scan period. This
along-track "staring" makes the scan
mirror's cross-track scans nearly perpendicular to the
spacecraft ground track, minimizing end-
of-scan overlap and gaps.
-
20
Chapter-3
Analysis of scanning motion
The Landsat thematic mapper acquires data during both forward
and reverse sweeps of its scan
mirror. The scan velocity is essentially constant during both
directions of scan for two reasons:
1) Torque is applied to the scan mirror only during the
turnaround times (not during the active
scans), 2) The scan mirror floats in a nearly inertia-free
manner during the active scan through
the use of magnets through the use of magnets which compensate
for the flex pivot spring forces.
.So the torque should be applied during turnaround times not
during active scans. So using
spherical geometry method, target latitude and longitude of
target point has been detected by
using the relationship between target, satellite and earth
center. The following sections gives
brief idea about this method and it is taken from the ref.
[2].
3.1 Spherical Geometry method: Spherical geometry method is used
to get target point location in terms of Latitude and Longitude.
The following fig.3.1shows the angular relation between satellite,
target and Earth center [7].
Figure 3.1 : Definition of angular relationships between
satellite, target and Earth center
The most common problem in space mission geometry is to
determine the relative geometry of
objects on the earth surface as seen from the spacecraft. One of
the methods is to use the given
-
21
coordinates of a target on the Earth to determine its
coordinates in the spacecraft field of view.
Another method is to determine the intercept point on the
surface of the Earth corresponding sub
satellite point of the spacecraft coordinates.
The spherical-Earth approximation is adequate for most mission
geometry applications. The
Earths oblateness has two distinct effects on the shape of the
Earth as seen from space. First, the
Earth appears somewhat oblate rather than round, and second, the
center of the visible oblate
Earth is displaced from the true geometric center of the
Earth.
The spacecraft is at a point P and the corresponding sub
satellite point on the surface of the
Earth is S with a distance r from the surface of the Earth. The
spacecraft sensor is pointing
towards E on the surface of the Earth with a distance d from the
surface of the Earth and tilt
angle T with respect to the sub satellite point .Elevation angle
, measured at the target between the spacecraft and the local
horizontal, Earth centered angle , measured at the Earth from the
sub satellite point to the target point as shown in the figure
3.1.
From law of sines: sin(90 ) sin sin
E
Tr d R
+ = =
From above relation cos sin sinE
r r Td R
= =
d in terms of is given by 2 2 2 2 cosE Ed R r rR = +
d can also written in terms of : sin cosdr
= 2
2 2cos 1 sin 1 cosdr
= =
then 2 2 2cos sinE Ed r R R = Tilt angle is given the relation
cos cos cosIT CTT = where IT = intrack angle; CT =crosstrack
angle
From fig. (3.1): cos sinE
r TR
= ; then cos sinEd r T R =
-
22
Arc on the ground is given by sin cos sinE
d d Tr R
= =
The Line of sight (LOS) vector which is pointing towards target
point and it is perpendicular to
the ground track of the spacecraft. The relation between unit
vector along the LOS ( lu ) and unit
vector along the spacecraft nadir ( nadiru ) is given by the
following section.
Figure 3.2: Ground track and sensor scan motion on the surface
of the Earth
Unit vector ( lu ) along = cos
sincos
IT CT
CT
CT
; nadir unit vector ( nadiru )=
001
[sin cos sin cos cos ]l nadir IT CT o CT o CT IT ou u i j k k =
+ = sin cos sinIT CT o CT oj i and l nadiru u = sin 1nlT ; where
1nl =unit vector along l nadiru u 2sin T = 2 2 2sin cos sinIT CT CT
+
This equation is equivalent to the equation cos cos cosIT CTT
=
-
23
Figure 3.3: Relation between different unit vector
From above relations sinl nadiru u
T
= 1nl = sin sin cossin sin
CT IT CTo oi jT T
+
Hence tan = sintan
IT
CT
The relation between , CT and IT is given as sintan tanIT
CT
= ;
where = angle between direction of motion of the spacecraft and
unit vector along l nadiru u From above relation, if IT =0, then
the angle = 0 and tilt angle is equal to cross-track angle.
Azimuth of the spacecraft orbit ( xaz ): 1tan
cos tanxAz
u i= where u=argument of latitude;
and i = inclination angle of spacecraft orbit
Spacecraft latitude ( ) : sin sin sinu i = ;
Inertial longitude ( ): cos sin sin cos costancos cos sin cos
sin
u u iu u i
+ =
Where is ascending node angle
-
24
Figure 3.4: Relation between sub-satellite coordinates and spot
coordinates of sensor (mirror)
From law of sines: sin sinsincos sin cos
s E
E
az az
= =
But from fig 3.3 it is shown that 2 x s
az az + + = ;
sin cos( )s xaz az= +
From triangle NSE spot latitude is given by: sin sin cos cos sin
sin( )E xaz = + +
For spot longitude: consider sin sincos sin
s
E
az
= ; hence sinsin sincos
s
E
az =
Relative longitude cos( )sin sincos
x
E
az + =
To determine target latitude and longitude of the scanning
motion, we should know the cross-
track and in-track pointing commands for scanning. The following
section gives a brief analysis
about cross-track and in-track commands for scanning.
-
25
3.2 Cross-track and In-track pointing commands for scanning:
Consider some inertial frame in the orbit plane, and true
anomaly angle () is measured from the frame.
Figure 3.5: Relation between geometry as viewed from spacecraft
and from the center of the
Earth
The satellite attitude is often expressed in a frame of
reference in which the local vertical is
normal to the Earths ellipsoid (geodetic reference), rather than
pointing to the Earths center
(geocentric reference).The current development is based on
geocentric reference. The following
flow chart shows the several steps involved in transformations
from inertial frame of reference to
line of sight (LOS) reference.
The first step in transforming to the Line of sight frame is to
rotate the inertial frame counter-
clock wise around the z-axis by the angle .This is achieved by
the transforming
-
26
cos sin 0sin cos 00 0 1
o
o
o
l lb bn n
= where l bn is inertial vector, o o ol b n - rtn vector
The next transformation involves rtn frame of reference to orbit
frame of reference,the relation is
given as
0 1 0 sin cos 00 0 1 = 0 0 11 0 0 cos sin 0
o o
o o
o
X l lY b bZ n n
=
o oX Y Z - orbit frame of reference
The third step in transforming to intermediate LOS frame of
reference is to rotate the orbit frame
counter-clock wise around the y-axis by the in-track angle IT
.The resultant transformation is given by
'
'
'
cos 0 sin0 1 0
sin 0 cos
o IT IT o
o o
o IT IT o
X XY YZ Z
=
= 1 00 1 0
0 1
IT o
o
IT o
XYZ
= sin cos cos sin 0
0 0 1sin cos cos sin 0
IT IT
IT IT
lbn
+ +
The final step in transforming to LOS frame of reference is to
rotate intermediate LOS frame
counterclock wise around the x-axis by the cross-track angle CT
. The resultant coordinates are
'
'
'
1 0 00 cos sin0 sin cos
l o
l CT CT o
l CT CT o
X XY YZ Z
=
Where l l lX Y Z - line of sight vector; ' ' 'o o oX Y Z -
intermediate LOS frame vector
-
27
I I IX Y Z l bn= (Inertial frame of reference)
IZ n = (Rotate inertial frame counter-clock wise around IZ axis
by the angle) o o ol b n (Similar to Radial, transverse, normal
(RTN) frame)
o o oX Y Z (orbit frame of reference) where ; ;o o o o o ol Z b
X n Y= = =
IT oY (Rotate orbit frame counter-clock wise around Y-axis by
the angle IT ) ' ' 'o o oX Y Z (Line of sight intermediate
frame)
'CT oX (Rotate intermediate frame counter-clock wise around
Y-axis by the Cross-track angle CT ) l l lX Y Z (Line of sight
frame of reference)
The line of sight vector in LOS frame 00
l
l
l
XYZ
=
So 0
[ ][ ] 0
LOSo
o IT CT
o
XYZ
= =
1 0 1 0 0 00 1 0 0 cos sin 0
0 1 0 sin cos
IT
CT CT
IT CT CT
o
o
o
XYZ
=
1 00 1 0
0 1
IT
IT
0sin
cosCT
CT
=
cossin
cos
IT CT
CT
CT
LOS vector is given by pr r = +
where r - vector from Earth center to spacecraft as shown in
fig.3.5 ; pr = vector from Earth
center to target point.
-
28
The scan motion is along normal direction (n ) at some
particular time, suppose at *t t= , then the
corresponding true anomaly angle is * ,then *
*
l lb bn n
=
The scanning is along n at a fixed vertical direction ( *l ) so
*
p E nr r l z n= + this relation is only at
the time of scanning only. pr =
* *
0
l b n
E
n
r
z
F
where Er = Earths radius; nz = scanning motion of
sensor with respect to Earth surface
Now the spacecraft orbit frame at *t t= , = * , then pr = * * *0
X Y Z
n
E
zr
F
To determine instantaneous orbit frame then transform the orbit
frame at *t t= , = * rotate it about z-axis by the angle * = . The
following flow chart shows the transforming to the instantaneous
orbit frame of reference
from satellite orbit frame at *t t= , = * , and also
transforming to the instantaneous rtn orbit frame from RTN frame at
*t t= , = * .
* *l b n (rtn frame at *t t= , = * ) * * *X Y Z (The satellite
orbit frame at *t t= , = * )
* n = * *Z = o ol b n (The instantaneous o o oX Y Z
(instantaneous orbit frame)
rtn orbit frame at t)
-
29
Figure 3.6: Relation between geometry of satellite motion and
scan motion
The vector pr in orbit frame is given by
* * [ ]cos sin 0sin cos 0 0
0 0 1
o
l b n
E
p
n
scanningrr
z
=
F
F
cossin
o ol b nE
E
n
rr
z
=
o ol b nE
E
n
rrz
(the instantaneous rtn orbit frame at t)
Transforming to instantaneous orbit frame, as we know the
relation RTN frame and orbit frame is ; ;o o o o o ol Z b X n Y= =
=
Hence o
prF =
X Y Zo o oE
n
E
rzr
F
(the instantaneous orbit frame)
-
30
Line of sight vector by using above relations pr r = +
= 00
o o o X Y Zo o oX Y ZE
n
E
rz
r r
+
F
=E
n
E
rz
r r
The above vectors is used only during scanning of the sensor
(mirror) that means it is instantaneously while scanning of the
mirror taking place and equate this vector to the satellite orbit
frame of reference. It is given by
pr r = + =E
n
E
rz
r r
=
cossin
cos
IT CT
CT
CT
From the above relation crosstrack angle is given by 1 1sin tann
nCTE
z zr r
= = and also
In-track angle is given by ;cos
EIT
CT
r =
To find angular velocity vector ( ) = 2
We know that pr r = + and also pr r = + * ( )p E n or r l z t t
n= + = *E nr l z n+ ;
Because * *l b n is a frozen frame and ignoring Earths rotation:
pr = nz n = n oz j where o o oi j k - Orbit frame unit vector
r = vb = ovi (v- velocity of the satellite)
So o n ov i z j = = 0
X Y Zo o o
n
vz
F
Transform this orbit frame vector to line of sight vector ( l l
lX Y Z ) frame: = [ ] [ ] ooIT X CT Y F
= 1 0 1 0 00 1 0 0 cos sin
0 1 0 sin cos 0
IT
CT CT n
IT CT CT
vz
= cos sin
sin cosn CT IT CT
n CT IT CT
vz vz v
-
31
Then = [... ... ... ]l l l lk i j k + + where li lj lk - unit
vector triad in the LOS frame
The angular velocities: , ,o IT CT ; where o - orbit angular
velocity , IT - in-track angular velocity , CT - cross-track
angular velocity = ' 'o o IT o CT lj j i + + where ' '[ , ]o o o lj
j i i= = = '( )o IT o CT lj i + + Transformation matrix from orbit
frame of reference to LOS frame of reference is
1 0 00 cos sin0 sin cos
CT CT
CT CT
, so 'oj =
0cossin
l
CT
CT
F
and li = 100
l
F
substitute these relations in
so = ( )o IT + 0
cossin
l
CT
CT
F
+ 00
l
CT
F
= ( ) cos( )sin
l
CT
o IT CT
o IT CT
+ +
F
but = 2
= 1 [ ( cos sin )]l l n CT IT CTvj i z v + +
by comparing above vector: CT = 1 [ cos sin ]n CT IT CTz v + and
IT = cos oCTv +
3.2.1 Extremities of intrack and crosstrack motions: The
analysis which is discussed so far is for coasting (scanning)
period, to determine turnaround
limits for the intrack commands it is better to know correction
for satellite motion in Local
vertical and Local Horizontal (LVLH) frame . Figure 3.7 shows
correction of satellite motion in
LVLH frame and th_it_mx = maximum in-track angle, th_ct_mx =
maximum crosstrack angle
Turnaround: in-track angle
At time t = ot initial conditions are - ,IT mx , ,IT mx and
final conditions ,IT mx , ,IT mx First half with positive
acceleration , then intrack rate commands are given by
, ( )IT IT mx IT ot t = +
-
32
Figure 3.7: correction for satellite motion
By integrating above in-track rate, then ( )IT t = 2, , 1( ) (
)2IT mx IT mx o IT ot t t t + During first half turnaround period (
a ) in-track commands are , ,IT a IT mx IT a = + ; ,IT a = , ,IT mx
IT mx a 212 IT a + The next half with negative acceleration (- IT )
,then in-track and intrack rate are
IT = ,IT a - IT (t - ot ) and IT = ,IT a + , ( )IT a ot t - 12
IT2( )ot t
for the next half turnaround period ( a ) then ,2IT a = ,IT a -
IT a = ,IT mx as desired
,2IT a = , ,IT a IT a a + - 212 IT a ,IT mx
,2IT a = , ,IT mx IT mx a 212 IT a + +( ,IT mx IT a + ) a -
21
2 IT a
= ,IT mx - 2 ,IT mx a + 2IT a ,IT mx
so 2IT a = , ,2( )IT mx IT mx a + then , ,22( )IT mx IT mx
aITa
+=
-
33
then ,IT a = ,IT mx + 2a ,IT mx
During coasting (scanning):
I
prF =
cos sin 0sin cos 0
0 0 1
o ol b nE
E
n
rrz
=
(cos sin )(sin cos )
I Il b nE
E
n
rr
z
+
Consider Ib component: (sin cos )Er = sin( )Er = *sinEr During
turnaround period pr = r +
oo o o
o
X Y ZE
n
E
rz
r r
= =
F
F Correct during scanning only, 00
o o oX Y Z
rr
=
o F =
cossin
cos
oo o oX Y Z
IT CT
CT
CT
=
F
this is correct for any IT , CT
then pr =cos
sincos
oo o oX Y Z
IT CT
CT
CT r
=
F
=coscos
sin
o ol b nCT
IT CT
CT
r
The next step is transforming to o ol b n orbit frame target
vector ( pr ) is to rotate the inertial frame counterclockwise
around the z-axis by the angle .This transformation is given by o
ol b npr = [ ]Z Ipr F But here to find inertial frame from o ol b n
orbit frame ,this can be done by
I
prF = [ ]Z o ol b npr
I
prF =
cos sin 0sin cos 0
0 0 1
coscos
sin
o ol b nCT
IT CT
CT
r
=
cos ( cos ) sin ( cos )sin( cos ) cos cos
sin
CT IT CT
CT IT CT
CT
rr
+
Consider the Ib component:
sin( cos ) cos cosCT IT CTr + = sin cos sin( )CT ITr + (for 1IT
) For small = ( )ITr altitude + = ( ) ITr h h +
-
34
3.3 Results: Figures 3.8, 3.9 shows the cross-track and in-track
results for scan analysis by using Landsat parameters which is
mentioned in section 2.4.1 by using data like turnaround time,
cross-track angle, orbit rate, spacecraft velocity.
3.3.1 In-track and cross-track Results:
An object-space scan mirror sweeps the TM line of sight back and
forth seven times each second
in a direction normal to the orbital ground track. Data
collected during forward and reverse scan
and scan velocity is essentially constant [15]. So in
cross-track profile at the time of scan period
the velocity of the mirror is constant and at the time of
turnaround velocity changes with respect
to time. The angle, rate, acceleration profiles are shown in
fig.3.8
Figure 3.8 : cross-track command
The cross-track angle command which is shown in the fig.3.8, we
can approximate as sinusoidal
with an amplitude 8.56 D [(7.7(coasting)+2*0.43(turnaround)] at
7Hz frequency and in-track
angle command can be approximate as saw tooth signal. However,
this approximation can be
use in the analysis of robust controller design which is
discussed in the further section.
0 0.2 0.4 0.6 0.8 1-10
-5
0
5
10
Time (s)
CT
(deg
)
crosstrack command
0 0.2 0.4 0.6 0.8 1-400
-200
0
200
400
Time (s)
rat
e CT
(deg
/s)
0 0.2 0.4 0.6 0.8 1-5
0
5x 104
Time (s)
acc
CT
(deg
/s2 )
-
35
Figure 3.9: In-track command
Figure 3.10: compensated motion with respect to LVLH frame
If we combine the in-track and cross-track motion, and feed it
to the scan mirror then the resultant motion
is simulated. The SLC correction for orbital motion to the
mirror with respect to the spacecraft (LVLH
frame) is shown in fig.2.8. The SLC jumps ahead during the scan
turnaround so that the next set of
0 0.2 0.4 0.6 0.8 1-0.02
-0.01
0
0.01
0.02
Time (s)
IT (d
eg)
intrack command
0 0.2 0.4 0.6 0.8 1-5
0
5
10
Time (s)
rat
e IT
(deg
/s)
0 0.2 0.4 0.6 0.8 1-2000
-1000
0
1000
2000
Time (s)
acc
IT (d
eg/s
2 )
-10 -8 -6 -4 -2 0 2 4 6 8 10-0.02
-0.015
-0.01
-0.005
0
0.005
0.01
0.015
0.02
CT (deg)
it (d
eg)
-
36
scan lines is contiguous and parallel with the previous set
which is compensated motion as
shown above fig.3.11.
The TM scans cross-track swaths of 187km as mentioned in section
2.4.1, downrange of the scan
motion with respect to normal to the orbit and sensor motion on
the surface of the Earth in km is
shown in fig.3.11.
Figure 3.11: Scan pattern of the scanning mirror on the surface
of the Earth
3.3.2 Analytical Results based on Spherical Geometry:
Implement the above in-track and cross-track commands in
spherical geometry method, so that
we can easily find out the coordinates of the scanning motion of
the sensor (scan mirror). The
orbit parameter of Landsat spacecraft are taken from [15]. But
the inclination angle of the
spacecraft is considered as 90 D instead of 98.2 D because it is
easier to analyze the motion of the
spacecraft and sensors, and rests of the parameters are remain
same.
-150 -100 -50 0 50 100 1500
2
4
6
8Scan Pattern
normal to orbit (km)
Dow
nran
ge (k
m)
0 0.2 0.4 0.6 0.8 1 1.2 1.4-200
-100
0
100
200
Time (s)
zn (k
m)
-
37
357.8 358 358.2 358.4 358.6 358.8 359 359.2 359.4 359.6
359.80
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Latit
ude(
deg)
Longitude(deg)
spacecraft subsatellite pointsub-satellite point starting
pointsensor scanning motionsensor's ground track starting point
Suppose if there is no in-track motion and only cross-track is
given to the mirror, then we can
expect the resultant motion as zig-zag manner as it mentioned in
section 2.4.3.The simulated
result is shown in fig. 3.12.
Figure 3.12: Un-compensated scan motion of the scanning
mirror
If the in-track angle is also include to the mirror motion then
the resultant motion has smooth
turnaround curves at the end of each active scan and also the
active scans are parallel to each
other for successive scans. This is happens on the surface of
the Earth, anyhow in this simulation
we neglected the Earth rotation because the simulation is
carried out for 1sec.If we observe this
corrected motion in local vertical and local horizontal frame
(spacecraft point of view) it is as
fig.3.10. That means in-track is not only corrects the motion at
turnarounds but also at the
(coasting) active scan. These results are expected and simulated
result is shown in the fig.3.13.
The variation of spacecraft latitude, longitude and sensor
latitude, longitude with respect to the
true anomaly has been found out and the results are in
Earth-fixed frame.
-
38
0 0.02 0.04 0.06 0.08357
358
359
360
true anomaly(deg)
spot
long
itude
(deg
)
0 0.02 0.04 0.06-0.05
0
0.05
0.1
true anomaly(deg)
spot
latit
ude(
deg)
0 0.02 0.04 0.06 0.080
0.05
0.1
true anomaly(deg)
latit
ude(
deg)
0 0.02 0.04 0.06358.761
358.762
358.763
358.764
true anomaly(deg)
Long
itude
(deg
)
Figure 3.13: Compensated scan motion of the scanning mirror
Figure 3.14: Relation between true anomaly Vs spacecraft and
sensor latitude and longitude
357.8 358 358.2 358.4 358.6 358.8 359 359.2 359.4 359.6
359.8-0.01
0
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
Latit
ude(
deg)
Longitude(deg)
satellite ground tracksatellite ground track starting
pointsensor ground tracksensor's ground track starting point
-
39
Figure 3.15: variation of spot distance with respect to time
Figure 3.16: variation of tilt and elevation angle with respect
to distance
0 0.2 0.4 0.6 0.8 1 1.2 1.4705
706
707
708
709
710
711
712
713
Time (sec)
Spo
t dis
tanc
e(K
m)
705 706 707 708 709 710 711 712 7130
5
10
Distance(Km)
Tilt
angl
e(de
g)
705 706 707 708 709 710 711 712 71380
85
90
Distance(Km)
Ele
vatio
n an
gle(
deg)
-
40
Figure 3.17: Variation of elevation angle with respect to tilt
angle
Figure 3.18: Variation of true anomaly with respect to
cross-track angle
0 1 2 3 4 5 6 7 8 981
82
83
84
85
86
87
88
89
90
Tilt angle of mirror(deg)
Ele
vatio
n an
gle(
deg)
-10 -8 -6 -4 -2 0 2 4 6 8 100
0.01
0.02
0.03
0.04
0.05
0.06
0.07
0.08
thetaCT (deg)
(Tru
e A
nom
aly,
deg
)
-
41
Figure 3.19: variation of sensor latitude and longitude with
respect to the spacecraft latitude
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08-0.05
0
0.05
0.1
Latitude of spacecraft(deg)
Latit
ude
of s
cani
ng m
irror
(deg
)
0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08357
358
359
360
Latitude of spacecraft(deg)Lon
gitu
de o
f sca
ning
mirr
or(d
eg)
-
42
Chapter 4
Controller Design Analysis for Scanning Mirror
4.1 Introduction:
A control system is a dynamic system that contains a controller
as an integral part. The device
that generates the signal according to control law and controls
the response of the plant is called
controller. The purpose of the controller is to generate control
signals, which will drive the
process to be controlled in the desired manner. Certain command
inputs given to the controller
and the plant is expected to behave in a desired manner, under
control. In feedback control, the
plant has to be monitored and its response needs to be measured
using sensors, for feeding back
into the controller. Then, the controller compares the sensed
signal with a desired response as
specified externally and uses the error to generate a proper
control signal.
There are different types of control systems, depend on the loop
of the system like open loop
control system, closed loop control system, feed-forward control
system. If the plant is stable and
completely, accurately known and if the inputs to the plant can
be precisely generated and
applied, then accurate control might be possible even without
feedback control. Under these
circumstances, a measurement is not needed and thus, we have an
open-loop control system. In a
feedback control system, the control loop has to be closed, by
measuring the system response and
employing that information to generate control signals to
correct any output errors. Hence,
feedback control is also known as closed-loop control. In
feedforward control, unknown inputs
are measured and that information, along with desired inputs, is
used to generate control signals
that can reduce errors due to these unknown inputs or variation
in them.
Figure 4.1: An open-loop control system
Controller Signal
conditioning ActuatorPlant
Set point
(Reference
input)
Controlled Variable (Output)
-
43
Reference Output input
Figure 4.2: A system with feedback and feedforward control
The current work dealt with different types of the controllers
like PD (Proportional-derivative),
PID (Proportional-Integral-Derivative), Double-lag compensator.
PD, PID controller design
comes under time-domain analysis, where as Double-Lag
compensator comes under frequency
domain analysis.
4.2 Time Domain Design Analysis:
The time-domain design of the control systems refer to the
utilization of the time-domain
properties and specifications of the system to be designed. The
time domain characteristics of a
linear control system are represented by the transient and
steady state responses of the system
when certain test signals are applied. In this case the in-track
and cross-track commands which are
simulated by using analysis (section-3.2) .Qualitatively, the
damping ratio and the natural
frequency can be used to indicate the relative stability of the
system.
4.2.1 PD controller Design:
With ideal derivative compensation, a pure differentiator is
added to the forward path of the
feedback control. Proportional derivative (PD) consists of
feeding the error (proportional) plus the
derivative of the error forward to the plant.
Controller Plant
Measurement for
feedback
Measurement for
feedforward
-
44
The transfer function of PD controller ( )cG s = KP + DK s;
The transfer function of the Plant ( )pG s = 21
Js; where J is mass moment of inertia
The closed-loop transfer with unit feedback of the overall
system is given by
( )1 ( ) ( )
c
c p
G sG s G s+ =
' '
2 ' '
KK
P D
D P
K ss K s
++ +
where 'K P = 1KPJ
; 'DK = 1
DJK;
Controller ( )cG s plant ( )pG s
Reference + controller output input (Actual output)
(-)
Figure 4.3: A closed-loop system with unit feedback PD
controller in forward path
The closed loop characteristic equation 1 ( ) ( ) c pG s G s+ =
2 ' 'KD Ps K s+ + ; (4.1)
The characteristic equation of standard form of the second
ordered system is given as
2 22 n ns s + + (ref.10) (4.2) The dynamic behavior of the
second-order system can then be described in terms of two
parameters , and n or c . If 0
-
45
Control frequency ( c ) = 8 scan frequency = 112 = 351.853
rad/sec By using equations 4.1, 4.2 'DK = 2 c , 'K P = 2c
'DK ,
'K P values are depends on , c , where as changes from 0.1 to 1
for a fixed c value. 'DK ,
'K P can get by using another method which is used in PID
controller design and it will be
discuss in further section.
c = 8 c = 9 c = 10 zeta 'K D
'K P 'K D
'K P 'K D
'K P 0.1 70.37184 123804.318 79.16832 156689.839 87.9648
193444.2460.2 140.74368 123804.318 158.33664 156689.839 175.9296
193444.2460.3 211.11552 123804.318 237.50496 156689.839 263.8944
193444.2460.4 281.48736 123804.318 316.67328 156689.839 351.8592
193444.2460.5 351.8592 123804.318 395.8416 156689.839 439.824
193444.2460.6 422.23104 123804.318 475.00992 156689.839 527.7888
193444.2460.7 492.60288 123804.318 554.17824 156689.839 615.7536
193444.2460.8 562.97472 123804.318 633.34656 156689.839 703.7184
193444.2460.9 633.34656 123804.318 712.51488 156689.839 791.6832
193444.2461 703.7184 123804.318 791.6832 156689.839 879.648
193444.246
Table 4.1: Variation of 'DK , 'K P as damping coefficient
changes for cross-track motion
Let consider damping ratio as 0.7 and c = 8 then substitute
these values in closed-loop transfer function is 2
351.859 492.6492.6 351.859
ss s
++ +
Loop transfer function is ' '
2
K P DK ss+ = 2351.859 492.6ss
+
It is mentioned in the ref. [15] that the geometric accuracy of
SMA for cross-track scan motion is
20 RAD (1.14millidegree). The time domain results shown in the
fig.4.6, that the cross-track error is more than the permissible
limit. The open loop and closed loop frequency response of the
PD controller is shown in the fig.4.4, 4.5. The phase margin of
the system is 73.5deg, and the gain
margin is infinity. The double integrator plant is stable for
infinite gain with PD controller.
-
46
.
Figure 4.4: PD controller Bode plots
-40
-20
0
20
40
60
80M
agnitu
de (d
B)
101
102
103
104
-180
-135
-90
Phas
e (d
eg)
crosstrack PD compensator bode plot
Frequency (rad/sec)
-20
0
20
40
60
80
Mag
nitu
de (d
B)
intrack PD compensator bode plot
Frequency (rad/sec)10
110
210
310
4-180
-135
-90
System: cPhase Margin (deg): 73.5Delay Margin (sec): 0.000777At
frequency (rad/sec): 1.65e+003Closed Loop Stable? No
Phas
e (d
eg)
-
47
Figure 4.5: closed loop frequency response for cross-track and
in-track PD controller
10-2 100 102 104 10610-10
10-8
10-6
10-4
10-2
100
102
(rad/sec)
Mag
nitu
decrosstrack closed-loop frequency response
magnitude=0.707
10-2 100 102 104 10610
-10
10-8
10-6
10-4
10-2
100
102
(rad/sec)
Mag
nitu
de
intrack closed-loop frequency response
magnitude=0.707
-
48
Figure 4.6: PD controller error plots for cross-track motion
(0.9-damping ratio)
0 0.2 0.4 0.6 0.8 1-0.4
-0.2
0
0.2
0.4
0.6
0.8
1
time(sec)
cros
stra
ck(d
eg)
crosstrack motion for 10Wcrosstrack motion for 8W
active scan
permissable error1milli degree
active scan
0.06 0.08 0.1 0.12 0.14 0.16 0.18 0.2 0.22 0.24
-0.15
-0.1
-0.05
0
0.05
0.1
time(sec)
cros
stra
ck(d
eg)
crosstrack motion for 10Wcrosstrack motion for 8W
active scan active scan
permissableerror 1millidegree
-
49
But in case of SLC (scan line corrector) which imparts in-track
motion to the mirror has different
specifications and scan mirror assembly parameters are not
valid. So the frequency of SLC is
14Hz, the