Self-driving cars, also known as autonomous vehicles, use a range of computerized navigation and control technologies to operate without a human driver. The terms semiautonomous or partially autonomous vehicles typically refer to vehicles equipped with technologies that can carry out some or most operational functions independently of driver input. In the 2010s, major technology corporations and automakers have spurred the development of autonomous vehicles that are powered with complex, intricate sensors and connected technologies.
In 2016 the Society of Automotive Engineers released official standards, which have been adopted by the US Department of Transportation (USDOT), outlining six levels of driving automation. Level 0 (assistive technology such as automatic emergency braking), Level 1 (steering or braking support such as lane centering or adaptive cruise control), and Level 2 (simultaneous steering and braking support) require a human driver to control the vehicle at all times. Level 3 technology provides limited automated driving but will quickly revert to human control if the system malfunctions. In Level 4 and Level 5, the technology can control all areas of driving in some or all conditions, respectively, and will come to a stop in the event of a system failure.
Despite major technological improvements, a large number of Americans remain skeptical or fearful of self-driving cars. An American Automobile Association (AAA) survey released in 2019 reported that 71 percent of Americans would be afraid to ride in a fully self-driving vehicle. Aggregated poll data from 2016, 2017, and 2018 published by the consumer safety organization Advocates for Highway and Auto Safety indicated that a large majority of Americans feel uneasy or unsafe about automated vehicles and would greatly prefer to ride in a vehicle driven by themselves or another human driver.
Self-Driving Cars in the Twenty-First Century
Two prominent companies involved in the development of modern autonomous vehicles include Alphabet, Inc., which operates its autonomous vehicle program under a subsidiary called Waymo, and Tesla Motors, led by entrepreneur Elon Musk (1971–).Other noteworthy industry stakeholders include rideshare giant Uber, technology corporations Nvidia and Apple, and major auto manufacturers such as GM, Nissan, and Mercedes-Benz.
Waymo, which began as a project under Alphabet subsidiary Google in 2006 before becoming a separate company in 2016, is widely considered the industry leader. According to PC Magazine, Waymo led all autonomous car companies in 2018 by covering more than 1.27 million miles in autonomous vehicle prototypes on California roads. Waymo representatives state that the company's autonomous vehicles are equipped with advanced sensors that provide computerized driving systems with 360-degree views and three-dimensional mapping capabilities. In May 2019 Waymo launched a pilot project in Phoenix, Arizona, where it introduced a fleet of ten Level 4 self-driving vehicles available to customers on the Lyft rideshare platform. The Lyft customers are accompanied by a rider from Waymo.
Regulating Autonomous and Semiautonomous Vehicles
As autonomous and semiautonomous vehicle technologies approach widespread implementation, governments are creating laws to regulate their safe use. In 2019 the National Highway Traffic Safety Administration (NHTSA) updated its Federal Automated Vehicles Policy recommendations for state-level governments and highway safety officials as the United States continues to move toward vehicular automation. The NHTSA's updated policy encourages the USDOT to provide technical assistance to prevent state governments from encumbering technological development with needless regulatory impediments. It also called for the standardization of autonomous vehicle terminologies, and for states to voluntarily begin conducting safety assessments on their road networks to determine their suitability for self-driving cars.
The House of Representatives passed the SELF DRIVE Act in 2017, an attempt to regulate autonomous vehicles at the federal level. The proposed legislation did not pass the Senate. In 2019, however, both Republican and Democratic senators expressed interest in revisiting such legislation. If passed, the act would establish universal technical standards for self-driving vehicles on public roadways, along with additional safety and cybersecurity requirements. States have also responded with their own measures: fifteen US jurisdictions passed a total of eighteen automated vehicle-related bills in 2018, and almost all states and the District of Columbia have either considered or passed legislation related to self-driving cars since 2012.
Safety and Security
Some experts believe that autonomous vehicle technologies hold the potential to drastically reduce traffic fatalities caused by driver distraction, driver impairment, and human error. However, while they can reliably detect and avoid problems associated with regular traffic flow and potential obstacles and impediments, current automated vehicle technologies have some issues reacting to erratic and unpredictable elements, such as the type of sudden movements often associated with children and elderly pedestrians.
Numerous high-profile incidents have influenced public opinion of the safety of self-driving cars. In June 2016 a motorist using Tesla's Level 2 autonomous driving system was involved in a fatal crash. The accident occurred on a divided highway after a large truck turned into the self-driving car's path. The self-driving car's sensors failed to detect the truck, and the driver was unable to react in time to avoid a crash. An investigation revealed that the driver did not have his hands on the wheel as mandated by Tesla safety guidelines, leading authorities to rule that the crash could have been avoided and was partially caused by human error. Another widely publicized incident took place in March 2018, when a pedestrian was struck and killed by a self-driving Uber vehicle while crossing the street with her bicycle in Arizona. As in the June 2016 incident, authorities ruled human error to be partially at fault, as the vehicle's safety operator, who was responsible for making emergency evasion maneuvers in such a situation, was distracted at the time of the crash.
Emerging safety concerns surround the security of the internet-connected systems that play a major role in operating autonomous vehicles. Cybersecurity experts note that no system connected to an external network is 100 percent invulnerable to hacking and outside intrusions. As such, it is theoretically possible for a cybercriminal to hack into a moving autonomous vehicle, assume control of its operation, and deliberately cause a high-speed or otherwise fatal crash. Hackers could also hold autonomous vehicles and their passengers for ransom—a particularly worrisome possibility for automated mass transit solutions like self-driving buses. Similar actions could see hacked vehicles arranged to create gridlock, thus shutting down a city's surface transportation network. Maintenance personnel could also alter or otherwise interfere with a vehicle's upkeep records to pressure its owner into paying for unnecessary repairs, updates, or upgrades.
Consumer Acceptance and Understanding
Consumer skepticism of fully autonomous vehicle technologies is well-documented, and market researchers also report that people generally distrust the assistive and semiautonomous technologies already available. A 2018 review published by Strategy Analytics found that while consumers have moderately favorable attitudes toward the automatic parking technologies currently offered on some production vehicles, they generally prefer to avoid vehicle features that put a computerized driver in control even for a limited period of time. The trend was especially pronounced among older survey participants and female drivers, and the report stated that large numbers of North American and Western European consumers say they do not trust self-driving car technologies and never will. A 2019 market survey by Deloitte posted similar conclusions while noting that consumers also heartily favor strong government oversight of automated vehicle technologies. Thus, for autonomous vehicle manufacturers, changing public perceptions of self-driving cars remains a significant and pressing challenge as the technology continues to become ever more market-ready.
Some industry experts assert that these beliefs reflect a predictable resistance to major technological innovations, similar to those who opposed automatic transmissions when they were first developed and introduced. Additional research insights have indicated that consumers have inaccurate impressions of autonomous vehicle technologies, with 70 percent of respondents to a 2018 Thatcham Research poll believing it is already possible to buy a fully autonomous passenger vehicle that requires no driver input whatsoever. In reality, only very limited forms of semiautonomous vehicle operation are available as of June 2019, and consumers generally overestimate their capabilities and incorrectly believe they can relax, watch programming, read, or otherwise stop paying attention to the road when using them.
Timelines that attempt to predict when autonomous vehicles will be ready for widespread distribution vary. Tesla founder Elon Musk has stated that by the middle of 2020, the company will have more than one million taxis with reliable Level 5 autonomous driving technology, though these predictions have largely been dismissed as exaggerated. Honda and Hyundai have stated that they will have Level 3 models capable of self-driving on highways ready for 2020, while Toyota has announced that it will be offering Level 2 technology through its Lexus brand the same year. Renault-Nissan claims it will have autonomous vehicles suitable for use on urban road networks by 2020, while multiple manufacturers project that truly driverless cars will become a reality by 2025.