Meraglim Blog

Autonomous Vehicles: The New Normal in the Near Future

There is a lot of buzz about driverless cars right now as automotive and technology companies have begun to roll out these autonomous vehicles. If you have been following the developments in self-driving cars, you are likely familiar with all the benefits that are touted by its advocates: fewer accidents, less congestion and pollution, improved land use, and more mobility for the disabled.¹ Many people are quite excited for this technology, while others remain skeptical and fearful of it. Regardless of which side of the spectrum you fall on, the technology is fascinating and presents many possible applications that are important to consider. It is likely that in the near future, autonomous vehicles will be a part of our daily lives, so it is best to gain an understanding of how this new technological development could impact you.

History and Development

The first attempt at a driverless car was in 1925 by a man named Francis P. Houdina.² Houdina toured the country that summer, demonstrating how his invention worked. The automobile, called the American Wonder, was controlled via radio by a driver who followed the driverless car in a separate vehicle. While Houdina continued to tour the car for several years, it had significant safety issues; at one point, it crashed into a car filled with cameramen. Therefore, it was never mass produced or sold commercially.

After this, technology focused more on developing automatic highways rather than automatic cars. In 1956, GM introduced the concept of the Firebird II, a car that would be steered with an electric highway.³ This idea was taken a step further in 1959 by the Radio Corporation of America. A test model for this “highway of the future” was built in Princeton, NJ, which featured an electrical cable that warned the car of any obstructions ahead, which then triggered the car to either brake or switch lanes. The engineer behind this project, Vladimir Zworykin, believed that the widespread use of this technology would address the growing problem of automobile accidents in America, and anticipated these electric highways to become the norm within a decade or two. However, the infrastructure costs both in installation and maintenance stopped the project before it really began. In 1977, Japan’s Tsukuba Mechanical Engineering Lab released what was considered the first truly autonomous car.⁴ The vehicle worked using analog computer technology and two cameras for signal processing. It was able to drive up to 18.6 MPH with the help of an elevated rail.

In the ‘80s, movies such as “Christine” and “Knight Rider” brought into the public eye the idea of an autonomous vehicle.⁵ Simultaneously, more research started being conducted into making driverless cars a reality. One research team at Bundeswehr University in Munich turned a van into a self-driving vehicle that they called the VaMoRs. At the same time, researchers at the Carnegie Mellon Robotics Institute started a line of robotic cards with the transformation of a panel van into an autonomous car. These projects required vans in order to hold all of the necessary equipment.

In 2004, the Defense Advanced Research Projects Agency (DARPA) unveiled the Grand Challenge series, which is a competition that gives robotics departments at universities and robotics companies the opportunity to develop an autonomous vehicle.⁶ While DARPA is clearly concerned with this technology for military purposes, the technology that is coming out of this project are directly creating the commercial driverless cars that are being developed. Google has hired the best researchers to come out of the Grand Challenge series to develop the Google car.

How They Work

Current models of self-driving cars build on several features that are already used in driver-controlled cars. For example, developments in the ‘80s and ‘90s in vehicles focused on automation to enhance safety.⁷ An example of this is anti-lock brakes; while previously, the driver needed to pump the brakes to prevent them from locking, anti-lock brakes do the pumping for you. About a decade of the development of anti-lock brakes, automobile manufacturers took this technology a step further to create sensors that adding traction and stability control. These steps in safety development of standard cars represents the move towards driverless cars. Additionally, the evolution of other technologies has contributed to the development of autonomous vehicles.

LiDAR

Light Detection and Ranging (LiDAR) is a technology that autonomous vehicles use to create a three-dimensional map and detect hazards by using laser beams to determine the distance between the car and other objects.⁸ In the Google Car, a Velodyne 64-beam laser is mounted on the top of the car so it has an unobstructed, 360-degree view on a custom-built, rotating base.

Radar

While LiDAR works to accurately map surroundings, it is unable to monitor speed of surrounding vehicles. Therefore, autonomous vehicles also use bumper-mounted radar units for this purpose.⁹ Driverless cars are outfitted with two sensors on the front bumper and two on the back bumper, which send signals to the on-board processors to move out of the way or apply the brakes. Radar technology works with other technologies in the car, such as gyroscopes, a wheel encoder, and inertial measurement units, in order to help the car make informed decisions to avoid crashes.

Cameras

The cameras used in the different models of driverless cars vary, but cameras are commonly used.¹⁰ One example is that it is mounted to the exterior, allowing the car to get an overlapping view of its surroundings. Just as the human eye uses different overlapping images for peripheral vision, depth, and dimensionality, the driverless car uses multiple cameras to create a complete image. The camera has a 50-degree view up to 30 meters. Cameras are just one piece of the puzzle that is a driverless car; if they were to malfunction, the car could still work.

Sonar

Some prototypes of autonomous vehicles also include sonar technology.¹¹ Sonar technology has some disadvantages when compared to LiDAR and radar technologies; namely, it has a narrow field of view and a short effective range. What sonar technology truly adds is a redundancy that allows the car to cross-reference multiple sources of data to make its decisions.

GPS

In order to get anywhere, autonomous vehicles require advanced positioning systems to keep on course and to route to its destination.¹² In the case of the Google Car, it uses Google Maps, GPS satellites, a wheel encoder, and inertial measurement units to maintain correct driving speed and route the car. Using real-time data from the cameras combined with GPS, the system can determine the speed of cars surrounding it, while correcting for things like construction or traffic.

Advanced Software

The software required to power driverless cars is quite impressive. Using all the data that it acquires from the sensors, cameras, and GPS, this software create algorithms to make its important decisions.¹³ The car is hardwired to respond in certain ways, such as stopping at a stop sign, but many responses are learned from experience. Every time a self-driving car is used, it learned more solutions that can be used in different situations, just like a human driver. This software also processes the data of other cars in order to learn.

Military Applications

Like many forms of technology, autonomous vehicles as we know them today developed out of military research. In 2016, the U.S. Army began testing self-driving vehicles, part of a long-term plan to implement driverless technology in the battlefield.¹⁴ First, they tested semi-autonomous trucks in Michigan, driving as a convoy over the course of seven miles.¹⁵ A three-phase pilot program is run by the Army Tank Automotive Research, Development and Engineering Center (TARFEC) out of Fort Bragg, NC.¹⁶ The vehicles are essentially autonomous golf carts which are used to pick up injured soldiers at the barracks and take them to the medical center, which is about a half-mile distance. One goal of this project is to cut military costs; as appointments at the medical center are quite expensive, if a soldier fails to show up, it is a huge cost. The hope is that by providing this transportation, they will cut down on missed appointments.

A more military important goal for this technology is to improve safety for soldiers.¹⁷ Driverless technology could enable the military to send supplies into dangerous areas without putting soldiers at risk. By removing the need to have soldiers drive convoy vehicles, they are removed from roads that potentially contain IEDs. Additionally, the implementation of driverless cars could speed the deployment process, allowing freight trucks to move at maximum efficiency in large platoons without the concern of human error. Additionally, the military uses autonomous helicopters to carry cargo to risky areas in which a human would be unable to safely and efficiently deploy.¹⁸ Between 2011 and 2014, these autonomous helicopters carried more than 4.5 pounds of cargo throughout thousands of missions in Afghanistan.

Commercial Applications

There are currently several automobile and technology companies developing self-driving cars. Last October, Tesla announced that it would begin equipping their cars with technology that would eventually allow them to become completely autonomous.¹⁹ GM has announced that within a year, it will be testing self-driving electric taxis with Lyft.²⁰ Uber currently has 100 driverless cars in use in Pittsburgh, which are free as they are only part of a test at the moment.²¹ Additionally, a Singapore-based company called nuTonomy is testing a self-driving taxi service. Other automobile companies that are currently looking into driverless cars include Ford, Volvo, Honda, and Fiat.²²

Everyone has heard of the Google Car, and currently, it looks as though it is well on its way to be commercialized; Google recently created a new company named Wayno, indicating that it is past the research phase and it now ready to release the car on the market.²³ Meanwhile, it seems Apple’s efforts to create their own self-driving vehicle is being downgraded in light of lack of progress and the success of other companies.²⁴ Clearly, the market of driverless cars is quickly developing, with some saying that they could be commonly used on the road by 2025. The eminence of this change is demonstrated by the United States Department of Transportation, which released the Federal Automated Vehicles Policy as a guideline for automakers.²⁵

What’s Next

Given the speed with which this technology is now developing, we can expect to see more and more autonomous vehicles come to the commercial market within the next 10 years.²⁶ Beyond cars, we will see many other autonomous transportation. Most intriguingly, Rolls-Royce is currently developing autonomous vessels under a project called Advanced Autonomous Waterborne Applications.²⁷ This has fascinating implications for the maritime industry. Robotic ships could make cargo shipping safer, more efficient, and less expensive.²⁸ Not only would it reduce accidents, it would also reduce the threat of piracy, because the ship could not be overtaken by any intruders. Additionally, with no need to accommodate a crew, these ships would have larger capacity for shipping, and make the ships lighter with fewer operating costs. This would also solve the issue of fewer and fewer people having maritime skills as seafaring becomes less and less attractive for work; instead, people with mechanical and electrical skills would be required more than people who could steer ships. Given how advantageous autonomous vessels could be, it is likely that by 2025, self-steering ships will be commercially used.

The technology behind autonomous vehicles will certainly have a dramatic impact in the coming years. In as little as a decade, we can expect the market to be saturated with self-driving cars. As with all new technology, there will be advantages and disadvantages, but one thing is for certain: autonomous vehicles will be revolutionary.