Lt. Bemotavicius is a Submarine Officer currently en-rolled in the USN Post Graduate School at Monterey, CA. This paper was recommended for publication by RADM Ray Jones, the current PGS USW Chairman.
Abstract
In this paper we examine the historical impact and future potentiality of computers on the art and practice of Undersea Warfare (USW), with particular emphasis on Submarine Warfare. While a complete history is impractical in this short format, a solid overview of the evolution of the use of computers from their nascent days, to mechanical prototypes and sophisticated modern systems is considered. Aside from considering the tangible benefits of computers, this paper also discusses the increasingly evident challenges that arc created by our ever broadening dependence on increasingly complicated systems. The conclusion is inescapable; computers provide not only an avenue to success, but also pose some of our deepest challenges.
Introduction
“Now if the estimates made in the temple before hostilities indicate victory it is because calculations show one’s strength to be superior to that of his enemy; if they indicate defeat, it is because calculations show that one is inferior. With many calculations, one can win; with few one cannot. How much less a chance of victory has one who makes none at all!” [Griffith, 1963] Though the concept of a submersible vehicle has existed for centuries, the modern use of a submarine as a submerged warship, and a capable weapon is a relatively recent innovation. Early submarines such as TURTLE and HUNLEY were little more than simple diving bells. Powered by hand and armed with only the most primitive of weapons-they served as little more than harbingers of a new type of warfare. Early innovations in submarines and submarine warfare were mechanical in nature. The development of newer and more efficient mechanical systems led the way for ballasting, ventilation, habitability, and of course weapons and sensors. These changes and improvements transitioned submarines from scientific curiosities to true weapons and warships.
Increasingly computers have played an important part, perhaps even the leading part, in the race to a better and more sophisticated weapon. Today computers dominate not only the day to day operation of the submarine, but also the planning of its missions, the training of its crew, the employment of its sensors, and the design of the boat itself. The submarine has given us a new vantage point from which to ask questions about the nature of our conflicts and their resolutions, and increasingly the answer to these questions is sought via technological means. At the same time computers also pose some of our strongest challenges, indeed they are changing the very notion of how our weapons evolve.
First Computers
Today, when we think of computers our tendency is to imagine screens and graphics, or perhaps a machine that can solve complex computations. In truth, the evolution of computers begins with devices such as a simple abacus. In similar fashion, the first computers used in undersea warfare were far from the complicated machines that we utilize today. One of the more useful early computing devices was the Torpedo Data Computer, or TDC. In fact, this device was actually much closer to the abacus than todays machines. Designed to solve the complex geometry problem of torpedo fire control, the TDC was an analog, electro mechanical device. The problem of aiming a torpedo at a moving target from a moving submarine was difficult and computationally intensive. One had to take into account the bearing to the target and its speed to calculate its advance, not to mention the torpedo speed and gyro angle, etc. Jn WWI this problem was addressed with a variety of slide rules [Wikipedia, 2008]. By WWII technical innovation had provided another solution. An analog device, the TDC was primitive in the sense of today’s computers, but sophisticated for its time and a real tactical advantage. It accepted inputs on the target solution from officers in the attack party, and provided an angle solver and an estimated target position. In some sense, it was the first true use of a computer in making tactical decisions. Certainly it was one of the first uses that permitted and aided real time decision making from the submarine.
The Navy’s use of computers rapidly progressed beyond simple mechanical devices such as the TDC. The age of the digital computer was upon us and though initial progress was slow, it would eventually come to dominate much of what we do. In former Secretary of the Navy John Lehman’s excellent book profiling heroic sailors and Ships, one of the few women detailed is RADM Grace Hopper, whom Lehman calls a “Naval Reformer”. [Lehman, 2001] In 1943 the graduate of Yale, and professor at Vassar joined the Navy WAVES where she was promptly assigned to work on programming America’s first digital computer-the Mark I. Fifty one feet long, it occupied a massive space and could perform only three additions a second. It was however the coming of an age for both the Navy and the nation. It was initially conceived of as a machine that might rapidly complete complex calculations-such as those involved in laying a minefield. [Lehman, 2001] RADM Hopper went on to work with increasingly complicated and increasingly smaller computers. She served in the Navy until the age of 80. She was perhaps one of the first to question why such devices could not be used for strike planning or navigation. Indeed, it is hard to imagine (perhaps impossible in the case of strike planning) doing any of these tasks without the aid of computers. This early use of computers was focused on a simpler or improved method to complete sophisticated calculations. It addressed problems that were already in existence, perhaps in a new and innovative fashion-but none-the-less they were problems that already existed. Today’s computers encompass not only large sophisticated machines shore side, but also a menagerie of such devices on board ships and submarines. From computers as an integral part of many systems such as sonar and fire control, to the desktop computers that arc now ubiquitous in the stateroom of every officer-it is clear that computers have come to dominate our technology, and in only 50 years! What is sometimes less clear is that often times the problems they aid with arc not even ones that can be conceived of without computer technology.
Operational Planning
Not only have computers come to dominate on board submarines, but also in the planning and analysis of submarine operations. As the US entered WWII, no comprehensive analysis had been done regarding the problem of an extended conflict featuring submarine warfare. Despite the lessons of WWI, namely that submarines can be a decisive advantage, no theory or plan had been develop d to cope with this threat. [Meigs, 2002)
Initially the scientists of the National Research Defense Committee focused on such problems as determination of the maximum range of echo-ranging and its environmental dependence, improvements in the problem of tracking a contact based on intermittent bearing and range data, and probability studies focused on devising optimal attack procedures. They also focused on gaining quantitative in formation about the actual result of attacks and operations. [Meigs, 2002) It is clear that these arc the sort of problems at which computers excel. They are both computationally intensive and data dependent. They arc also highly dependent on repetition-something that is difficult for humans to do efficiently. Indeed computer databases were used, though again of a primitive sort compared to todays fine computers. Initial databases were kept on so-called “Hollerith codes”, punch cards based on an automated codes.[Meigs, 2002) This sort of cumbersome and crude data would allow scientists to determine optimal tactics such as convoying and perfect attack patterns for aircraft and surface ships. It also signaled an entire field that would later arise namely that of computer databases and data-mi11i11g techniques. Today we expect large repositories of data we desire, whether it be a bibliography and electronic references, a collection of personnel records, or a large accumulation of tactical scenarios. In fact, many of our plans and assumptions rely on large depositories of data.
The ability to maintain large stores of data and conduct complex calculations, and then repeat them many times are the hallmarks of all computers. These are also the sort of abilities that allowed scientists and Naval officers in conjunction to invent the idea of operational analysis and to consider the quantitative optimization of tactics and analysis. In WWII computers also played a key role in cryptography and signals analysis.
Today computers are increasingly shifting operational planning from shore to sea based platforms-often with real time capabilities. Complicated strike missions can now be completely planned and executed from a submerged platform. What once took hours or days of planning on a mainframe can now be done in a few seconds in the control room of a submarine.
Training
Computers revolutionized USW in another, completely separate fashion-training. Training is perhaps one of the most significant factors in deciding any conflict-tactical or strategic. In order to be useful, training must be realistic, evaluated, and optimal. That is to say, one must train using the highest standards and ideal tactics with constant observations and measurable performance. It is hard to replicate the complexity of a tactical environment in such a setting, and yet the quality of training is completely dependent on it. Early attempts were by and large improvisation …
“At first thought, a simulated approach and attack might seem too nebulous to have real training value. However the same section of the TDC used to determine enemy course and speed could also generate a complete, realistic problem. To have all the element for a sonar approach required only the recording of the enemy ship’s speed as well as its range and bearing for each minute of the exercise. To introduce the sound bearings realistically, we developed our own device, consisting of my shaving brush and a dynamic microphone. The microphone was plugged into the receptacle in the forward torpedo room that normally received the output from one of our sound heads … ” [O’Kane, 1977]
Training rapidly proceeded from an improvisational use of the TDC and a shaving brush to a more complicated use of computer resources. Advanced trainers are particularly important in ASW, where the problems are often long-on the scale of many hours and filled with boredom. Here few problems will be seen by any crew (aircraft or surface) and the tactical behavior must be optimal. As early as 1943 scientists had developed suitcase sized training aids that could be used to simulate attack problems, as well as simulated sonar problems. A drastic improvement over an analog machine and a shaving brush!
Today modern trainers dominate preparation cycles for crews of all platforms. These can range from mock ups of a sonar space with real time problems, to virtual reality ship control trainers, the flight simulators. Increasingly, trainers feature connectivity that allows personnel working on separate tasks to communicate and interact in a fashion that emulates closer and closer the intricacies of a tactical situation. On board Virginia class submarines, full tactical scenarios can be run right in the control room. These feature sonar, radar, and ESM contacts as well as full use of all tracking systems. In fact the use of a photonics mast (an inherently computerized system) permits computer graphics to simulate visual contacts. That the technology could have evolved to this degree of realism is a testament to the infinite versatility of the modem computer
Acoustic Propagation and Sensors
As previously stated, one of the primary goals for scientists as WWII entered its peak was to quantify the range and capability of acoustic sensors. [Meigs, 2002] Acoustic sensors have proved reliable and enduring due to the strength of sound’s ability to travel and persist in water. Developing useful sensors has proved to be dependent on understanding and quantifying the environment. This combines oceanographic study with the need to exploit the power of computing. Early efforts were focused on shore side understanding of acoustic arrays and the basics of underwater acoustic propagation. Later efforts focused on development of optimal acoustic sensors and tactics.
The realization that the ocean environment controls acoustic propagation was profound for the future of sonar. The study of this idea promoted the use of physics, oceanography, and perhaps most importantly computers. The basics of any model of acoustic propagation relics on strong models of the physics of sound propagation in the ocean, but the implementation involves complex calculations and a multitude of repetitions. These arc precisely the traits that computers arc ideal for and excel at. Shore side acoustic models run on mainframes have evolved into smaller more compact programs that can be run on computers located on a submarine. Today’s generation of computers and programs, such as PCIMA T and STOA, can take real time data and provide accurate acoustic models-or at least as accurate as the data that is provided. The improvement is staggering-from a fundamental concept to a practical and tangible advantage. It is important that we recognize that our ability to compute and maintain this sort of accurate modeling is a real and tangible advantage-as valuable as a new submarine or weapons systems. At the same time these models can lead us down a terrible path. Often times with a machine to tell us the sound propagation paths and intensities we absolve ourselves of the need to educate ourselves on the environment and in situ analysis. The tendency is to believe the computer vice calculate and analyze, to rely on a mean vice the actuality of here and now. How much easier is it to operate a terminal than to learn the environment and the reality of the here and now?
The application of computers to sensors has not been limited to sonar. Computers permit ever more complex signal processing, allowing ever smaller signals to be extracted and analyzed from the surrounding noise. Radar, ESM, and MAD are all sensors that exist in their current fashion almost entirely due to computers. The shift from analog systems to digital implementations has allowed incredible advances in capability and size. The smaller, far more powerful sensors that exist today arc a tangible product of a few decades of research. WWII saw the advent of radar and ESM, today we have radars capable of detecting periscopes from miles away and ESM suites that can use pattern matching algorithms to classify signals. It is unclear where a limit will be found, (if there is one!) but it is also certain that the path there will be paved with circuit boards.
Modern Advantages
We have seen that the historical impact of computers on USW has been profound, but the impact is not just historical. Advances continue to be made in all of the fields mentioned. Our acoustic models continue to improve and become more sophisticated and closer to real time. Our planning tools have improved and shifted toward sea based, in-situ planning. In the design of our trainers we increasingly take advantage of computing power to provide more realistic, connected, and mission optimal trainers. Today we can train for specific missions with specific anomalies-all while being observed and graded. What is interesting though, is to examine some of the more modern applications of computers that are not simple derivatives of WWII technology.
Unmanned vehicles are becoming increasingly important in today’s tactical environment. They are a platform in which the entire crew has been replaced by a computer. They are able then to enjoy the advantage of small size, maneuverability, and low risk that this arrangement conveys. Early unmanned vehicles were little more than large remote controlled cars, but the evolution has been rapid. Today’s vehicles are small, dynamic, and highly autonomous. Capable of entering denied areas without risk to humans, unmanned vehicles arc perhaps the future of the Navy. This has promoted research in an incredibly diverse array of fields, such as path finding and optimal search, mechanics and dynamics, control systems, and power systems. Already, today’s vehicles display incredible longevity and capability. All of these areas are facilitated and dominated by computer methods. It is clear that to maintain an advantage in unmanned vehicles, the Navy will need to maintain a firm advantage in computing.
The increased power and reliability of computers has permitted their use in more mundane, but no less important applications. Our navigation and ship control systems are increasingly turned over to computers. Today’s ships rely on GPS and computer based navigation systems. Our submarines rely on automated depth keeping and steering. These factors work to reduce crew size and permit greater multi-tasking. Even the back-up systems are now more and more reliant on computers. Simple commercial programs such as Excel and Power Point permit clearer presentations and more sophisticated data analysis on the deck plate level. In a fundamental way the nature of day to day work on board a ship is changing. Increasingly, less time is devoted to the mundane tasks of ship control and seamanship, and more is devoted to data analysis and dynamic improvements-at least in concept! It is a hard and constant challenge to avoid simply turning over the elements of seamanship to computers.
Even the construction of our ships is being revolutionized by computers. Previously ships were designed on paper with conventional drafting techniques-and a great deal of actual design was still needed during the construction process. Today computer programs such as CA TIA arc used as industry standards. These programs permit three dimensional design and collision analysis on desktop computers. This innovation has streamlined the construc-tion process and permitted greater accuracy and uniformity in the construction process. It is clear that the impact of computers is felt in ways that early pioneers could not imagine.
Modern Challenges
The previous sections have extolled the virtues of computers in almost every facet of US W. It must be noted that the ever increas-ing prevalence of computers is not without some negatives. As computers become increasingly sophisticated, so docs the informa-tion that they require-perhaps leading us outside the realm we arc accustomed to think in. Today’s acoustic models can accurately handle dynamic, littoral environments, but to operate them requires detailed knowledge of oceanography and acoustics. Our tactical decision aids can use probabilistic considerations and optimization techniques to aid in search problems, but the operation of these systems and the interpretation of their output often requires sophisticated understanding of probability and search theory. Modern sonar systems arc capable of incredible signal processing and advanced algorithms for analysis and estimation-but these techniques arc often outside the scope of our sonar operators training. A II of these issues raise concerns with traditional training pipelines. Increasingly a mismatch can be seen between machine capabilities, and machine use. Often operators do not even under-stand what these systems arc capable of-much less how to provide the detailed and sophisticated information they require. Trouble-shooting and maintaining these systems is increasingly difficult. Although proposed, the Submarine Force docs not even have an IT rating! In an era of increasing networks, expectations for connectiv-ity and strong desire for data to be available from all sensors, for data fusion-it is hard to sec how these desires can be fulfilled with our current force structure.
Computers also have the potential to narrow our view point instead of broadening it. Today it is inconceivable that a presenta-tion not be given on power point, that our data not be collated in Excel. The potential exists that our officer core will be reduced to middle managers and technocrats, losing our edge as tacticians and leaders. It is increasingly cumbersome to maintain a lead in training in both fields. Today’s officers need not only understand all of the systems and the increasingly complex theory that underlies them, but also the overriding tactical needs. It is possible, that as comput-ers become ever more complex and the systems ever more dy-namic, that these two goals will prove incompatible.
Conclusion
In this p aper we have surveyed some of the incredible historical impacts that computers have had on the art and practice of USW. It is clear that computers have proved to be the revolutionary catalyst for many of our technologies and ideas. It is also clear that many of the contemporary advances arc being driven by more sophisticated and sma ller processors. Some of the most dynamic technologies and growing fields like, unmanned vehicles, ship design, and navigation, arc dominated by these advances. At the same time computers pose a bold challenge for our personnel wilh regard to training and operation. The ever more evident mismatch between capability and use should prove an alarming warning that we arc loosing an edge. The short time frame in which computers have been in use and become dominant should also provide a warning. How much longer can we expect the Navy to resemble its traditional image? The only thing that is clear is that continued research and training in computational fields must be a crucial and leading component of the Navy’ s future.
