Cassini case study

V. Hunter Adams (vha3), MAE 4160/5160, Spring 2020

from IPython.display import Latex
from IPython.display import Image
from IPython.core.display import HTML

In This Lecture . . .

1. Why go to space?

   • Both the spoken and the unspoken reasons.

2. Why is going to space difficult and expensive?
3. Why does this course exist?
4. An introduction to many of the topics covered in this course via a case study.

Why go to space?

This seems like the right question with which to start this class.

Exploration

1. Colonization, expansion of the limits of humanity, tourism
2. Science (astrobiology, geology, physics, etc.)

Look down

1. Weather forecasting/science
2. Surveillance (military and commercial)

Look out

1. Astrophysics, atmospheric sounding

No gravity

1. Microgravity science
2. Materials processing

Resources

1. Rare elements (gold, platinum, etc.)
2. In-situ resource utilization
3. Water

If you read grant applications, you will find all of the justifications listed above for sending something into space. If you talk to the people that write these applications, however, you will very soon come to learn that there are much more personal and much more difficult-to-articulate reasons to go to space. These reasons often and with a dot dot dot (. . .). Examples include:

1. There's just something about blank places on maps . . .
2. There's something about frontiers . . .
3. There's something about new worlds . . .

I've had the good fortune to work with some people that are as accomplished in this field as a person can be. If you spend more than an hour with these people, you'll soon start hearing things like those listed above. You'll hear the same things from students in this field. However, it seems to me that people are generally embarrassed to say these things. I don't know exactly why that is, but I suspect that it is because these motivations are very hard to justify, and smart people don't like to say things that they can't justify.

I'm saying these things at the start of this class to assure you that you don't need to be embarrassed about those feelings in this class. They are shared by me, and by your classmates.

I'll mention also that if you are trying to articulate some of these feelings (i.e. if you're trying to find the words to explain these feelings), go to literature. This is why literature exists. Good literature and good art articulates those feelings which you can't find the way to express. So, if you're trying to explain your interests in the unknown to yourself in words, you might consider seeking out literature about exploration and the unknown. My favorite in this category is Moby Dick. There's a particular line in that book that articulated a feeling that I'd had for a long time, perhaps it will do the same for you.

This is spoken by Ishmael, explaining why goes to sea:

I have an everlasting itch for things remote. I long to sail forbidden seas, and to land on barbarous shores.

Why is going to space difficult and expensive?

You need a lot of kinetic energy

In order to get into space, you must go up at least a few hundred km, and then you must accelerate to ~7.5 km/sec. We use rockets in order to obtain that energy, but rockets are incredibly inefficient.

A tremendous amount of propellant is required in order to achieve the required $\Delta V$. Most space systems end up being close to 90% propellant by mass, which means that most of the energy (propellant) is spend accelerating propellant.

Then, once you're in space, surviving there is hard! Because this has historically been such an expensive destination that requires such specialized equipment, it has become characterized by small markets, unique parts, and labor-intensive processes. This might be beginning to change with the advent of smallsats, but space is still a tough place to get to.

This course

Why this course exists

This class is called Spacecraft Technology and Systems Architecture. It might alternatively be titled "How to Design a Spacecraft," since that's essentially what we'll be discussing throughout this semester.

This course exists because the spacecraft design process, and spacecraft themselves, are both sufficiently different from other engineered systems to warrant separate consideration. These differences are a consequence of the environment in which spacecraft and spacecraft design take place. By "environment" I mean both the literal space environment, which is unique, and the political and funding environment, which is also unique. Both of these factors lead to machines which not only look and behave differently from terrestrial machines, but that must be designed via processes which are specific to spacecraft.

Subsystems

Of course, there are plenty of ways in which spacecraft are similar to other engineered systems. Like nearly every other complex system, spacecraft are considered in terms of subsystems. We divide spacecraft into smaller systems, each of which with a specific set of responsibilities. The power subsystem is responsible for gathering and distributing power throughout the spacecraft, telemetry and command is responsible for maintaining a communication link with ground operators, attitude control is responsible for orienting the spacecraft, etc. We similarly divide cars, human bodies, computers, and other complex systems into subsystems. This makes designing and considering the collective system more manageable.

Much of this class will be spent walking through each of these subsystems. We'll consider the spacecraft one subsystem at a time until we've considered each individual component of the collective. Specifically, we'll consider power, communications, thermal, structures, attitude determination and control, guidance and navigation, avionics, flight software, ground segments, and launch segments. We'll also consider some additional subsystems required for specific mission types, like life support and entry descent and landing.

For each of these subsystems, I want to do the following. If there is a fundamental equation, or a set of fundamental equations, at the core of that subsystem, then we'll go over that equation. For communications it's the Shannon-Hartley Theorem, for propulsion it's the Rocket Equation, for ADCS it's Euler's rigid-body equations, etc. We'll take a look at the first principles for that subsystem. Then, I'll discuss conventional strategies for meeting the responsibilities of each subsystem. We'll discuss reaction wheels, thrusters, radio communication, etc. Then, because this is a university, we'll discuss some of the more cutting edge strategies that we can expect to become more common in the coming decades.

Goals

Perhaps obviously, based on the topics I've just listed, this is a breadth class as opposed to a depth class. For each of the subsystems that I just listed, one could take a semester-long course (or more) and read multiple textbooks on associated topics. The goal of this course is not to become expert in any particular subsystem, but instead to understand how each subsystem interfaces with the others, the responsibilities of each subsystem, and how each subsystem fulfills those responsibilities.

In summary, the course objectives are:

1. To understand space missions and systems, and how the space environment and mission requirements drive spacecraft design.
2. To understand the fundamentals of spacecraft subsystems, including propulsion, attitude determination and control, power, structures, thermal, communications, and command and data handling.
3. To understand typical practices for designing space systems in a contemporary context of US commercial space and government agencies.
4. To simulate a spacecraft in operation at the level of a Preliminary Design Review (PDR) using state of the art tools, and identify and characterize subsystems for a preliminary spacecraft design.

The project

When learning abstract concepts, I've always found that it's good to have a concrete case study to which those concepts may be applied. With that in mind, this course involves a semester-long project with milestone due dates throughout the semester. This project is described in the syllabus, and I'll talk about it in some more detail at the beginning of the next class, after you've all had the opportunity to read about it. In short, you'll be working in small groups to design a spacecraft for one of five possible missions. The hope is that as we discuss each topic or subsystem, you'll be listening with your particular mission in the back of your mind.

The rest of this lecture

For the rest of this lecture, I want to introduce many of the topics that we'll cover throughout this semester through a case study. I am going to consider an actual spacecraft, and we'll walk through the requirements and subsystem-level design for that spacecraft. I'm not only doing this to give you a sense for what this class will be about, though that is part of it. The other reason is that I think it's extremely important to start a class like this with the big picture.

Many of you have likely had the experience of working on a complicated task, getting buried in some minutiae, and losing track of the big picture. Have you experienced that "wait, what am I even doing?" moment? As you go into the weeds to figure out some technical aspect of a complex problem, it can be easy to forget the larger problem that you're trying to solve. That's a problem that you learn to avoid by stepping back now and then and reminding yourself of the larger goals. This is important to do for yourself, in order to keep yourself motivated and working on the right problems. It's also extremely important to do as an engineering manager. You'll find that as soon as an engineer loses track of why what he or she is doing is important, and how it fits into the larger picture, the productivity and motivation of that engineer plummets. Many of you will end up managing engineering projects, and you'll find that much of your time is spent reminding engineers how their tasks fit into the larger goal.

With that in mind, I want for us to start with the larger goal. Then, as we get into minutiae throughout the semester, we'll have this big picture step back and remind ourselves of. We'll do a couple more of these case studies throughout the semester. Hopefully stepping back like this gives us all some immediate context for why each of the topics that we'll discuss is important. With all of that being said, let's consider our spacecraft of interest: Cassini.

Cassini case study

I chose Cassini for its familiarity, and because it's an example of an extreme success. I think you're all likely familiar with Cassini, so I won't spend much time introducing the mission. In short, Cassini was a mission to Saturn and the moons of Saturn. It launched on October 15, 1997 and last contact was September 15, 2017, almost 20 years later. In that time, it did the following:

1. Gathered 635 GB of data
2. Discovered 6 moons
3. Took over 450,000 pictures, many of which are as breathtakingly beautiful as they are scientifically valuable
4. Led to the publication of almost 4000 scientific papers

It was, by all measures, a radical success. It was a success because of its design, and the process by which it was designed. Let us take a look at that process and that design. And then, because I can't help myself, we'll take a few minutes to look at some of the photographs that Cassini sent home.

Requirements

The design of a spacecraft starts with very careful articulation of the requirements for that spacecraft. The requirements come first, before any aspect of the spacecraft itself is even considered. Then the spacecraft is designed in order to meet those requirements. We can draw a loose analogy to test-driven development in software engineering. In test-driven development, you write a series of unit tests for your software before you write the software itself. Then, you write the software and see if it passes the unit tests. When it passes all unit tests, then you know that your software is as complete as your unit tests are thorough.

In much the same way, we articulate a series of requirements, and then we consider our spacecraft design complete when it meets all of those requirements. The requirements come first. Spacecraft are pure function. Every aspect of their design and their appearance is a pure consequence of meeting requirements put in place before any aspect of the spacecraft was designed. The Hubble looks like the Hubble so that it can point very accurately and very quietly, and so it can accomodate a large optical payload. Voyager looks like Voyager in order to meet its own requirements. And Cassini looks like Cassini because of the requirements that it was designed to meet.

There's a subtle point buried in this. Spacecraft are always means to an end, there are never the end in and of themselves. Every spacecraft (to date) is launched in order to perform some task of scientific, commercial, or military value. The spacecraft design is entirely in the service of that goal, nothing else. This is true of a lot of other engineered systems, but by no means all of them. Consider cars, for example. Clearly some aspects of the car are ends in themselves. If cars strictly existed in order to get us from A to B, I argue that they would look quite different. There's an artistic element of some cars that enters into the equation, that doesn't enter into the equation for spacecraft. It's interesting to think about what spacecraft might look like if such things did factor into their design.

We're going to spend an entire lecture discussing how to write these requirements. Requirements are derived from stakeholder objectives. In the case of Cassini, there were many stakeholders. Many of the requirements came from stakeholders that were scientists, however. These objectives may be a bit fuzzy, but requirements are:

1. Unambiguous
2. Isolated
3. Concise
4. Measurable
5. Unique
6. Consistent

Here are some examples from Cassini's Phase A report. This report contains a very large number of scientific objectives, a few examples are given below:

1. Determine temperature field, cloud properties and composition of the atmosphere of Saturn.
2. Map composition and size distribution of ring material (Saturn).
3. Study the diurnal variations and magnetic control of the ionosphere of Saturn.
4. Etc.

Cassini's Phase A report includes a long list of scientific objectives at Saturn, the icy moons, the Jupiter flyby, Titan, and cruise. These objectives, you could say, are the point of the mission. From these objectives, we derive a series of very specific requirements for the spacecraft to meet. Again, the Phase A report contains a long list of these, I've only included a handful below as examples.

1. The orbiter shall provide telecommunications o and from 34m and 70m DSN [Deep Space Network] stations at multiple data rates up to 115.2 kbps at the asteroid and 67.9 kbps at Saturn, and shall provide onboard storage for at least $3.6 \times 10^9$ bits of data for use during cruise, probe mission, and the tour.
2. The Orbiter shall target and deliver the [Huygens] Probe such that its asymptotic velocity with respect to Titan will be below 6.8 km/s.
3. The Orbiter shall be designed to operate at distances ranging from 0.86AU to 9.2AU from the Sun and up to 10.2 AU from the Earth.
4. The Orbiter shall be designed to withstand 52 $R_J$ Jupiter flyby, Saturnian and ring plane crossing environments in the clear zones, sparsely populated regions and upper fringes of Titan's atmosphere.

And many, many more. This is how Cassini's design began. We begin with objectives, turn those into requirements, and design our spacecraft based on those requirements. With that in mind, let's take a look at Cassini's design.

Image("Cassini.jpg", width=800)

Power

Question: What is missing from the above diagram, that you usually find on Earth-orbiting spacecraft?

Answer: Solar panels. Why aren't there any solar panels? Recall the above requirement, which stated that the spacecraft must be able to operate up to 9.2 AU from the Sun. Recall also that power generated from a solar cell depends on the efficiency of that cell (how well does it convert solar energy into electrical energy?), and on the solar flux. At 9.2 AU, the solar flux is almost 85 times less than at Earth. I haven't told you how much power the various scientific payloads require, but suffice it to say that some very large solar cells would be required to generate adequate power.

So what does Cassini use as an alternative? RTG's, (Radioisotope Thermoelectric Generators). Cassini carried 33 kg of Plutonium-238, a highly radioactive alpha-emitter with a halflife of 87.7 years. Cassini's generators converted the heat generated by the decay of Plutonium-238 to electrical power. A lot of power. Even at the end of its life, almost 20 years after launch, Cassini's RTG's were capable of continuous generation of 600-700 Watts of power.

Image("rtg.jpg", width=500)

Above: one of the RTG's on Cassini. Below: a hot pellet of plutonium, the heart of the above RTG.

Image("Plutonium.jpg", width=500)

For power peaks, Cassini also carried a couple Ni-Cd batteries. For maneuvers that required short-term power draw that exceeded that generated by the RTG's, Cassini could pull from the batteries.

Nuclear power is not uncommon for deep-space missions, because solar flux is so low. New Horizons carried an RTG (actually, one of the spares from Cassini). Voyager carried an RTG, as did Galileo and Ulyses. The curiosity rover is also powered by an RTG. Spirit and Opportunity were not.

ADCS

Attitude Control

Among the requirements for attitude determination and control for Cassini is the following (Lee, Allan, and Gene Hanover. "Cassini spacecraft attitude control system flight performance." AIAA Guidance, Navigation, and Control Conference and Exhibit. 2005.)

Image("adcs.png", width=500)

Given these required pointing accuracies, and given that the required slew rate is quite slow (17.5 mrad/sec), we can make some informed guesses about the composition of Cassini's ADCS subsystem.

Question: Do you think that Cassini was 3-axis stabilized? Why?

Answer: Yes! Our pointing requirements require that each rotational degree of freedom be controlled (we can't be spinning about one axis). What sorts of actuators do you expect were used for Cassini's attitude control?

Answer: How do we make a spacecraft rotate? We either induce an external torque on the spacecraft via something like a thruster or a torque coil. Or! We use momentum control. We can store angular momentum inside the spacecraft and move that momentum between the internal actuators and the spacecraft body. This sort of momentum control is achieved with reaction wheels and control-moment gyroscopes.

We're going to spend some time discussing the tradeoffs associated with various forms of momentum control. But you may recall that CMG's give the spacecraft a great deal of agility, at the cost of expense and complication. Reaction wheels are comparatively much simpler and easier to control, at the cost of slew rate. Since Cassini need not slew quickly, it used reaction wheels. Their arrangement is shown below.

Image("rws.png", width=500)

Cassini carried four reaction wheels. 3 of them were "strapped down" in fixed orientations. These orientations were chosen such that all of their angular momentum vectors spanned R3, giving full 3-axis control of the spacecraft. The fourth reaction wheel was redundant and was attached to a movable platform. If any of the fixed reaction wheels failed, the fourth could be rotated such that it aligned with the failed reaction wheel and take its place.

As the spacecraft orbits Saturn, it is acted upon by external torques. The small amount of atmosphere will exert a torque, solar pressure will exert a torque, and the magnetic field will exert a torque. In order to compensate for these torques, the reaction wheels will need to store more and more momentum. Eventually, they will be saturated and we will need to dump that momentum via some external torque (generated either via torque coil or thruster). In Cassini's case, thrusters were used to dump momentum from the reaction wheels.

Attitude Determination

Recall from 4060 that attitude determination requires at least two vectors that are known in the spacecraft body frame and in the inertial frame. In deep space around Saturn, what sorts of attitude determination sensors do you expect for a spacecraft to employ?

Answer: Star trackers (x2, redundant), sun sensors (x2, redundant), and a fiber-optic gyroscope (x2, redundant).

The information from these sensors is fed into an estimator, which combines all of the measurements in order to arrive at an estimate for spacecraft attitude. We'll spend a lecture discussing these estimators.

Propulsion

Cassini's main engine (for trajectory and orbit changes) used bipropellant (N2H2 and N2O4). These bipropellants were stored in two large conispherical tanks with a total capacity of 3450 kg. This gives a total delta-V capability of approximately 2040 meters per second. It's worth noting that the total delta-V that Cassini achieved was much, much greater than this, thanks to strategic use of gravitational assists. Cassini carried two main engines (a prime and a backup) and 16 monopropellant hydrazine thrusters (8 prime and 8 backup).

An accelerometer was placed parallel to the spacecraft's Z-axis in order to measure the delta-V imparted to the spacecraft during a burn of the main engine. Before the start of each burn, this accelerometer would be calibrated for 1 minute. The burn would terminate when the target delta-V was achieved.

Question: What does bipropellant mean?

Answer: Bipropellant means two propellants. One is an oxidizer and the other is a fuel. The two chemicals are combined in a combustion chamber where they react, and then are ejected out of the nozzle. This produces thrust.

Question: Why are propellant tanks generally spherical?

Answer: A sphere has greater volume than any other solid with the same surface area. So, for a given volume, the sphere is the shape with the lowest mass requirement.

Image("engine.jpg", width=500)

Reaction thrusters are fueled by hydrazine monopropellant pressurized by helium. These thrusters are used for slewing and momentum dumping from the reaction wheels.

Guidance and Navigation

Like many deep-space missions, Cassini's location and trajectory was tracked from Earth, not autonomously on the spacecraft itself. This was accomplished using the Deep Space Network, a network of spacecraft communication facilities located in California, Madrid, and Canberra. These facilities are capable of tracking deep-space spacecrafts' positions and velocities by examining Doppler shifts on radio transmissions (velocity) and delta-differential one-way ranging (position).

Image("gold.png", width=500)

We'll spend some time discussing various strategies for guidance and navigation near Earth and in deep space. As spacecraft become more computationally capable, this particular subsystem is changing rapidly.

Communications

Cassini features a 3.67 meter high-gain X-band antenna, and two low gain x-band antennas mounted above the high-gain antenna. The high gain antenna is a rebuild of that which was designed for Voyager. Cassini uses a Reed-Solomon encoder to achieve a biet error rate of $10^{-6}$ at a downlink rate of 27.6 kbps at maximum range and 67.9 kbps at minimum range (to the 70 m DSN ground station).

The low-gain antenna was used during the cruise phase, and in the event of an emergency if the HGA pointing is lost.

We will spend two lectures on communications, and we'll discuss radio encoding methods in some detail.

Image("antenna.jpg", width=500)

Command and Data Handling

The command and data handling subsystem is responsible for moving information throughout the spacecraft. This includes uplink, downlink, subsystem intercommunication, and sequence execution functions. This subsystem distributes engineering data of the spacecraft attitude and rate, the time, and the state of the scientific payloads. All science and engineering telemetry is packetized.

Cassini used separate communications buses for command and telemetry, low data rate instruments, and high data rate instruments. Each bus was redundant (two buses for each purpose). Cassini carried two digital tape recorders for storing information between downlink passes. Each of these could store 125 MB of information.

All subsystems used the same microprocessor, the Sandia SA3300 which was radiation and single-event upset hardened. Cassini was programmed in C.

Thermal

Mireles, Virgil, and Glenn T. Tsuyuki. "A summary of the Cassini system-level thermal balance test: Engineering subsystems." SAE transactions (1997): 940-953.

Thermal control was a challenge for Cassini because of the range of distances that it would be traveling from the Sun. Even so, it used mostly passive means for controlling temperature.

The high-gain antenna acted as a global sunshade. During early flybys of Venus and Earth, the spacecraft could be oriented such that the antenna shadowed the electronics. The antenna itself was covered with a thermal paint and thermally isolated from the rest of the spacecraft, to the extent to which that was possible. Heat-sensitive components (like the magnetometer) remained stowed until the spacecraft was permanently beyond 0.97 AU.

Cassini did carry and employ electrical heaters, but their use was minimized by taking advantage of waste heat coming from the RTG's. Cassini used louvers as a low-power method for maintaining thermal control through large variations in heat loads, and was covered exhaustively in thermal blanketing.

Image("thermal.png", width=500)

Orbit

Getting to Saturn

In order to get to the Saturn system, Cassini first did a flyby of Venus, boosting it 7 km/sec. This was in December of 1998. In June, 1999, Cassini performed a second flyby of Venus. This flyby required slowing the spacecraft by nearly 450 m/s by firing the main rocket engine for 90 minutes. Cassini went directy into a flyby of Earth, gaining 5.5 km/sec. In January, 2000, Cassini did a flyby of Asteroid 2685 Masursky, and on December 30, 2000 it performed a flyby of Jupiter. After this flyby, Cassini coasted to Saturn and arrived in February of 2004.

(gif from Wikipedia)

Once at Saturn

Cassini spent the next 13 years orbiting Saturn and performing flybys of its various moons. It visited Titan (and dropped the Huygen's probe into its atmosphere), Rhea, Iapetus, Dione, Tethys, Enceladus, Mimas, Hyperion, Phoebe, Janus, Epimetheus, Prometheus, Pandora, Helene, Atlas, Pan, Telesto, Calypso, and Methone. It's primary mission was complete 3 years after arriving at Saturn, after which it had three extended missions. These included Equinox (2 years, 62 days), Solstice (6 years, 205 days), and Finale (4 months, 24 days).

(gif from Wikipedia)

Consequences

What did we get in return for Cassini's careful design? 13 years of science at Saturn. Plus the science that was achieved en route to Saturn, which included a flybys of Jupiter, an asteroid, and Venus. Cassini discovered new moons, determined the length of a day on Saturn, discovered lakes on Titan, discovered water in the plumes of Enceladus, and generally rewrote the textbooks on Saturn and its system of moons.

In Cassini's case, the most compelling way to communicate its success is with the photos that it took. I could spend hours showing you photos. This is just a handful, no more or less staggering than any of the others that I could have chosen. Go look at all these photos.

True-color mosaic of Jupiter, the most detail global portrait ever produced.

Image("jupiter.jpg", width=500)

Jupiter and Io.

Image("jupiterio.jpg", width=500)

Portrait of Saturn

Image("portrait.jpg", width=500)

Saturn and Mimas

Image("mimas.jpg", width=500)

Saturn and Titan

Image("titan.jpg", width=500)

Plumes of Enceladus

Image("plumes.jpg", width=500)

Glint off lake on Titan

Image("Titan_reflection.jpg", width=500)

Rings

Image("rings.jpg", width=500)

The Day the Earth Smiled.

As far as I'm concerned, the below photo deserves to be mentioned in the same breath as any of mankind's artistic achievements. It's rare for something to be as beautiful as it is scientifically valuable. This composit photograph is an example of such a thing.

Saturn is eclipsing the Sun, which has backlit the rings. Earth is visible to the lower right of Saturn.

Image("smile.jpg", width=900)

Image("annotated.jpg", width=900)

End of life

After more than a decade at Saturn, and nearly 20 years in space, Cassini began to run out of fuel. Scientists and engineers did not want to risk collision with and contamination of any of the moons that may harbor life, so they decided that the safest way to end the mission would be to crash Cassini into Saturn. In April 2017, the final phase of the mission began, which NASA called The Grand Finale. I'll let them tell you about it.

Isn't it interesting how emotional that is? This is one of the most interesting aspects, to me, of spacecraft engineering. Some of these missions are just dripping with drama. It's worth really thinking about why the end of Cassini's mission was such an emotional event, even for people that had nothing to do with the mission and little interest in space. Here is the last image that it sent home:

Image("farewell.png", width=500)

There's something about the fact that there was only one Cassini. Thousands of engineers and technicians worked for years and years in order to build that machine, and only that machine. Their effort gets wrapped up in the machine in ways that it simply doesn't for mass-produced items.

And then we sent it into the unknown. I don't have an emotional response when a low-Earth weather satellite dies. The folks at Planet Labs probably don't feel too sad when a dove dies, and the folks at SpaceX likely won't feel too badly if one of their Starlink spacecraft dies. These flagship missions of exploration are different, they're tapping into something else inside of us. I'm a fan of the history of exploration and I often read books about the early explorers (Cook, Magellan, etc.). I wonder how similar our feelings toward Cassini are to the feelings that the people of their time had toward them, when they finally returned from a long voyage with new maps of the world.

Learn to identify the planets in the sky. Learn where Mars, Saturn, Jupiter, and the rest can be found. It's a very interesting feeling to stand outside on a dark night and think about the fact that there are manmade objects at these impossibly far places.

What did we cover today?

1. Why we go to space.
2. Why going to space is difficult and expensive.
3. Why this class exists.
4. An introduction to many of the topics covered in this class via a case study.

Before next time!

1. Read the syllabus, come with questions!
2. Read today's lecture supplement.

Sources and further reading

1. NASA-TM-103374, CASSINI. Report on the Phase A study: Saturn Orbiter and Titan probe. https://ntrs.nasa.gov/search.jsp?R=19910008874
2. Lee, Allan, and Gene Hanover. "Cassini spacecraft attitude control system flight performance." AIAA Guidance, Navigation, and Control Conference and Exhibit. 2005.
3. Mireles, Virgil, and Glenn T. Tsuyuki. "A summary of the Cassini system-level thermal balance test: Engineering subsystems." SAE transactions (1997): 940-953.
4. Cassini planetary images: NASA.gov