Navigation on the shoulders of giants

The long, winding, technological road to GPS in every car

GPS at your fingertips may now be ubiquitous, but it took a convergence of tech over decades.

Larry Printz – Jun 24, 2020 7:30 am | 121

The world’s first modern in-car with an integrated GPS navigation system as seen in the Eunos Cosmo, Mazda’s Japanese-market upscale marque. Credit: Mazda

Have you ever gotten lost while driving? Probably not recently thanks to GPS navigation.

The touchscreen navigation/infotainment system in every new car or truck that some colloquially refer to as “the GPS” is technology that we take for granted today. In fact, it’s so ubiquitous, it’s hard to imagine new vehicles without them (though such vehicles did exist—people used maps and turntables). But the road to today’s navigation touchscreens is a winding one marked by a trio of evolving technologies that developed over decades. It took a while before these innovations came together to ultimately doom the humble—not to mention difficult to read and refold—road map.

The next time you don’t know how to get somewhere and have to rely on maps embedded on your car’s display, you can thank the atomic clock, the satellite constellations powering actual global positioning systems, and the humble touchscreen.

What time is it, really?

Of the three technologies that led to today’s GPS inescapability, the atomic clock might be the most unexpected. But it is crucial to GPS technology. Each GPS satellite contains multiple atomic clocks that calculate the time for each GPS signal within 100 billionths of a second. This allows banks to locate the time and place of the ATM that you used to deposit a check, allowing for timestamping precision in all financial transactions. It allows the Federal Aviation Administration to precisely track hazardous weather using its network of Doppler Weather Radars. It lets your cellular provider share its limited radio spectrum more efficiently so you can always place your call. And it ensures digital broadcasters that all songs arrive on the same station at the same time regardless of where you are.

(And, no, despite the name, atomic clocks are not radioactive. Thanks for asking.)

Like a traditional clock, atomic clocks track time using oscillation. A traditional clock counts the ticks created by the oscillations of a pendulum. A mechanical wristwatch uses the energy from a wound mainspring passed through a series of gears to a balance wheel, which oscillates back and forth. A digital clock uses the oscillations of a quartz crystal or the oscillations from the power line. Regardless of technology, they all use oscillation as a way to track time, and so too does an atomic clock.

An atomic clock employs an electric oscillator regulated by an atom’s natural oscillation movement between the positive charge on the nucleus and the surrounding electron cloud. This oscillation never varies, so unlike traditional clocks, the oscillator’s frequency can be used for extraordinarily exact timekeeping.

The idea was developed by Columbia University physics professor Isidor Rabi in 1945 using a technique he developed in the 1930s called atomic beam magnetic resonance. This could accurately measure the magnetic properties of atomic nuclei by detecting single states of rotation of atoms and molecules, and the technique in turn proved feasible as a way to precisely tell time. The first clock using atomic beam magnetic resonance was introduced in 1949 by the National Bureau of Standards (now the National Institute of Standards and Technology, or NIST) using ammonia atoms, but it wasn’t accurate enough. In 1952, cesium was found to be the most exact element, and that was used for the first time on a clock named NBS-1. Seven years later, NBS-1 went into service as NIST’s primary timekeeper.

By the time the 13th General Conference on Weights and Measures was held in 1967, an international standard was established: a second of time was defined as the 9,192,631,770 cycles of radiation that it takes for a cesium atom to vibrate.

For the first time, the world’s timekeeping was no longer based on astronomy.

A cesium atomic clock is still in use today for the US government’s official time. And this concept of the atomic clock would prove essential in the development of global positioning systems during the Sputnik era—it all has to do with synchronicity, which is essential when developing GPS.

Transit, the world's first global satellite navigation system, which was launch in 1960. — Transit, the world’s first global satellite navigation system, which was launched in 1960. Credit: DARPA

Where in the world?

Like so many new technologies, the idea of GPS was already in use even if we wouldn’t recognize it as such.

By 1924, right around the time that Rand McNally printed the inaugural “Rand McNally’s Auto Chum,” sailors started using technology based on early radios. Called radionavigation, it allowed navigators to pinpoint their ship’s location by contacting stations along shorelines and harbors. Once a ship requested assistance, multiple shore stations would determine where the ship’s signal came from and convey it to the vessel, helping the ship’s navigator to fix their position.

GPS systems outside the US

Certainly, the government’s development of the atomic clock and global positioning systems led to the creation of the commercial market, which soon developed smaller, faster, and less expensive devices, which in turn helped the military. Today, the US Air Force Space Command manages the GPS System, which has been developed with the participation of NATO countries and Australia, among others. That’s what’s used by your car’s infotainment system to provide navigation and satellite radio broadcasts.

However, the rest of the world has caught up, beginning with GLONASS, or Global Navigation Satellite System, a space-based navigation system implemented by Russia. Plans for it began in 1968, and the country finalized this system by 1976. By 1991, 12 satellites were in use, and operational two years later, although the overall system wasn’t fully functioning until December 1995. By 2002, the system was barely working. With help from the Indian Space Research Organization, India’s space agency, the Russian Space Agency got GLONASS fully functioning by May 2007.

Not to be outdone is Galileo, the European Space Agency’s global navigation satellite system, one that operates in concert with GPS and GLONASS. The first satellites were launched in 2011. The system became operable in 2013, and now it has 26 satellites in orbit. When complete, Galileo will have 30 satellites in use.

Finally, there’s Beidou satellite navigation system, originally designed for the Chinese military. Beidou allows China’s military to no longer be reliant on the US-owned GPS for military and civilian operations.

This laid the groundwork for modern satellite-dependent global positioning systems, which started once the Soviets launched the Sputnik satellite. Scientists at MIT realized they could gauge the satellite’s distance using the Doppler Effect, measuring the radio signals’ frequency and distance from earthbound receivers by measuring the shifts in its radio signal to learn of the satellite’s distance. The Russian satellite’s radio signals increased as it advanced and decreased as it departed, not unlike the sound of a car’s engine (its pitch changes as it approaches, passes, and leaves).

This led the United States Navy to build its first satellite-based navigation system in 1959. Called Transit, it was developed by the Johns Hopkins Applied Physics Laboratory using orbiting satellites to send radio signals to monitoring stations. This allowed users to determine their position by measuring the Doppler shift of the satellites’ signals and in turn was used to track nuclear submarine locations. Although communication wasn’t fast, it proved the reliability of space-based systems.

Around the same time, the Naval Research Laboratory and the Space and Missile Systems Organization of the US Air Force had different ideas. They preferred the Timation program, created in 1964 and launched in 1967, which employed two experimental satellites that used quartz-crystal clocks (although they eventually switched to rubidium and cesium atomic clocks). And, as if to prove the existence of government inefficiency, the Air Force was also working on a similar technology program, called System 621B. This continuously provided navigation using 16 satellites in orbits that formed four oval-shaped clusters extending 30 degrees north and south of the equator. (Did we mention that the Army proposed its own system, SECOR, or Sequential Correlation of Range?)

Finally, someone at the Department of Defense formed NAVSEG, or the Navigation Satellite Executive Group, in 1968, charged with looking at the various systems already in place or being developed. Not surprisingly, the solution that emerged used the best aspects of the Navy and Air Force systems. Dubbed the NAVSTAR Global Positioning System, the DoD approved its development in December 1973. Testing began the following year, and full-scale development was approved in August 1979.

NAVSTAR hit some snags as the new decade dawned, unfortunately. Its development budget was cut 30 percent by the Secretary of Defense in the early 1980s. A new tank or plane is easy to understand; the importance of a new high-tech support system seemed harder to appreciate. The number of satellites was ultimately cut to 18 from 24, plus three spares. Then, just as it was about to proceed, the Space Shuttle Challenger accident in 1986 delayed the program another 24 months. The first satellites finally launched from Cape Canaveral in February 1989 and went into use in April, and not a minute too soon.

The 1990-91 Persian Gulf War proved the importance of GPS in combat. Along with infra-red night vision, this technology arguably helped win the war. GPS enhanced precision bombing and also allowed for the positioning of troops and certain special forces operations. The success of GPS in the Persian Gulf War uncovered its commercial possibilities, and by 1993, civilian use of GPS began in cars through third-party software, offered free of charge.

Digitized maps migrate to the instrument panel

The idea that GPS embedded in your car could tell you where to go, as opposed to your partner or ride co-pilot, was a far-fetched idea until the first map-based car navigation system was introduced in Japan as a dealer option for the second-generation Accord in 1981. Called the Honda Electro Gyrocator, it employed sensors and gyroscopes that compared the road surface to that of a map, thus guiding the car. Known as dead reckoning, the system didn’t use satellites. But the technique didn’t always work since maps aren’t always accurate. That’s a problem for a system that adds an additional 25 percent to the cost of the car. It was the first modern navigation system, but sometimes, being first doesn’t ensure longevity. The Honda Electro Gyrocator disappeared for 1982.

The makers of Etak know about that whole first but not last thing, too. Developed in California by Stan Honey and Atari founder Nolan Bushnell, this aftermarket automotive navigation system also operated on dead reckoning by comparing a car’s location to points on a map. The unit was designed by Pong designer Alan Alcorn, and it used an Intel 8088 microprocessor and cassette tape drive for the mapping. Through the use of a digital map, a compass, and wheel sensors, drivers would see the system’s results on a screen.

Designing the hardware was easy; getting the map information proved more challenging before Etak tapped into the expertise of the US Census Bureau, who were pros at digitizing maps using topographic mathematics. For a storage medium, engineers decided to store the maps on cassette tapes with polycarbonate shells. Each one held 3.5MB of data. Better yet, they resisted shock and vibrations and could withstand the extreme temperatures of a parked car. Nevertheless, the first area they mapped, San Francisco, required drivers to use six cassettes. The system’s screen was a vector-based Cathode Ray Tube that was high-res enough to display a bright, crisp line. A bitmap display was considered too pricey.

In July 1985, the Etak Navigator arrived as the 700 with a 7-inch screen for $1,595 ($3,819 when adjusted for inflation), and the 450 with a 4.5-inch screen for $1,395 (or $3,340 today). Cassettes cost $35 each (or $84 today). It was successful enough that future navigation competitors licensed Etak’s patents, map data, and/or hardware. Etak was bought by Rupert Murdoch’s News Corporation in 1989 for nearly $25 million.

Two years later, Toyota’s Japanese-market Crown Royal Saloon G offered an in-dash navigation system with the first color CRT display and CD-ROM-based mapping. But since all of these systems all used dead reckoning—which, again, compares a car’s location to that of a map—these solutions weren’t actually that much more advanced than the 1909 Jones Live-Map. There was just more technology involved in the 1980s.

It took Mazda to introduce the world’s first modern in-car GPS navigation system—which the company offered only in Japan on the 1990 Eunos Cosmo. Eunos was Mazda’s upscale marque, and the vehicle was slated to become part of Mazda’s planned Amati brand. Mazda ultimately decided against launching it in America, but nevertheless, the car was produced through 1995.

One year after the Cosmo, in 1991, Toyota introduced its “Electro-Multivision, Global Positioning System” on the Japanese market 1991 Toyota Soarer (known Stateside as the Lexus SC). It displayed the car’s location on a 6-inch color LCD screen with the aid of GPS satellites.

Soon, some of the names you might remember from GPS systems of yore started to materialize. In 1992, General Motors’ Oldsmobile and Delco divisions introduced a built-in GPS navigation system with a 6-inch full-color Sony CRT named TravTek on Avis Rent-A-Car vehicles in Florida. It eventually became a $1,995 factory option on the 1995 Oldsmobile 88 sedan, where it was known as GuideStar, a dead-reckoning navigation system using maps and not to be confused with the later OnStar service, which did not use maps. Initially only offered with maps of California or Las Vegas, the screen plugged into the middle of the instrument panel and could swivel left or right depending on who was navigating. A GPS signal and CD-ROM-based maps supplied the driving directions.

And in 1997, Japanese company Alpine introduced an aftermarket system that also used CD-ROM-based maps and GPS signals, allowing any car buyer to add it to their car. The following year, Garmin introduced its first portable StreetPilot GPS navigation system, and Garmin eventually rose to prominence in the aftermarket, cross-brand compatible market.

Touch me babe

In-car navigation rapidly spread throughout the auto industry in the 1990s. But few of these solutions were easy to interact with since they lacked one more modern tech touch: the touchscreen.

The humble touchscreen itself dates back to 1965, and it was born out of work of the advanced research arm of Britain’s Air Ministry.

E.A. Johnson invented the world’s first capacitive touchscreen, which is composed of multiple layers of glass and plastic and coated with a conductive material like indium tin oxide or copper. When you touch the screen, an electric circuit is completed at that point, commanding the operating system to respond. A capacitive touchscreen either senses it or it doesn’t; you must use a finger to complete the circuit. Johnson’s touchscreen would be employed by British air traffic controllers through the 1990s. It would also see use in millions, if not billions, of smartphones, laptops, and tablets due to this approach’s longer durability, superior multi-touch performance, and clarity compared to reactive touchscreens, which first came about in 1970.

The reactive touchscreen was invented by Dr. G. Samuel Hurst while studying atomic physics at the University of Kentucky. Moving on to the Oak Ridge National Laboratory, he continued working on his discovery. Hurst realized that when a finger touches a screen composed of numerous thin resistive layers with thin gaps between them, a computer can read the location of the resulting voltage and actuate the command. Not only could it be actuated with a finger, glove, or stylus, it was less expensive to build than a capacitive touchscreen. This screen is now commonly used on ATM machines and checkout line point of sale terminals because of its affordability.

Despite this technology being around for more than a decade, the first touchscreens for consumers didn’t arrive until 1982 in the form of the Hewlett-Packard HP-150, a personal computer run using an MS-DOS operating system on a 9-inch Sony touch-sensitive CRT. Being new technology, it was not cheap: $2,795, or $7,634 when adjusted for inflation.

Soon after, the world got the car that pioneered touchscreen use: the 1986 Buick Riviera. Frankly, it’s a vehicle that has been overlooked—mostly due to its small size and lackluster styling. But Buick’s first use of a touchscreen as standard equipment in the 1986 Riviera was actually a trailblazing triumph considering it took more than five years to develop the car.

According to company documents: in November 1980 (two years before the Hewlett-Packard HP-150 went on sale), Buick managers in Flint, Michigan, decided they would produce a car with the industry’s most advanced electronics by 1985. As a committee assessed which electronic features would be offered, a touch-sensitive CRT was being independently developed by General Motors’ subsidiary Delco Systems across the country in Santa Barbara, California. Within a few months, in early 1981, GM’s system was demonstrated to and approved by GM’s Product Policy Group. AC Spark Plug and Delco Electronics developed the hardware, while the Delco Systems handled the software design. By 1983, the specifications for the screen were decided and installed in a test fleet of 100 Rivieras the following year to measure customer reaction.

Dubbed the Graphic Control Center, the GCC was a CRT screen covered by an invisible Mylar panel that used transparent conductors that were row and column encoded to perform a specific function on a particular page. Each switch’s function changed with each page. Since CRTs take a few seconds to warm up, the GCC’s circuitry warmed up when the driver’s door handle was touched. By the time the driver’s door opened and closed, the display came on, revealing the Riviera logo.

Once the car was started, the display went to its home page, which handled 90 percent of a driver’s needs. If the screen wasn’t touched within 30 seconds, it shut off. The GCC controlled the automatic climate control, AM/FM radio, graphic equalizer, and trip calculations while also displaying gauges and vehicle diagnostic information. Its black screen and green display now seem like a relic, but this was cutting-edge at the time.

Buick executives were enthralled, including Cary Wilson, who first studied the idea in 1980. “A new generation of automobile electrical systems is at hand, and Buick has set the stage,” he said in 1986.

Others, such as legendary automotive journalist Brock Yates, were less sanguine.

“The reality of the Graphic Control Center is a bad joke,” he wrote in September 1986. “The Riviera’s setup does nothing that a conventional array of knobs, buttons and analog instruments could not do in a fraction of the time.” Colleague Rich Seppos agreed. “The Riviera’s high-tech CRT isn’t an advance, it’s a handful.”

Despite the negative reaction, Buick would install the GCC in the 1988-89 Reatta as well before a modified version appeared: Oldsmobile’s Visual Information Center, which was available in the 1989-92 Oldsmobile Toronado Trofeo (the Riviera shared its underlying platform). The 4-inch full-color touchscreen was made by Sony, and the system could be fitted with an optional Motorola cellphone that could be controlled through the screen.

While critics derided these and other early attempts at touchscreen systems, new cars such as the Tesla Model 3 come with all of their controls on a touchscreen and nowhere else. So, though all of this technology seems new, the science fueling it has been decades in the making—today’s Tesla console has been built on the shoulders of technological giants. The atomic clock, and its extreme chronologic accuracy, paved the way for the creation of the global positioning system using space satellites. Although originally designed for military use, those satellite systems eventually opened to the commercial market. And when paired with easy user-facing technologies such as touchscreens, the market for what we now know as GPS blossomed.

Our current ability to track where we need to go through a touch screen is ubiquitous—but never forget it took decades of development, and a convergence of technologies, to get here.

Larry Printz is an automotive journalist based in South Florida. He can be reached at TheDrivingPrintz@gmail.com.

121 Comments