7 Things You Might Not Know About Autonomous Vehicles
Automakers are developing ever-increasing levels of automation in their vehicles and are rolling out a dazzling array of new features geared toward a future of self-driving cars. This includes everything from automatic steering, braking, and acceleration to AI-based systems that use sensors and cameras to ‘read’ the environment around them and react accordingly. As the highly automated driving industry races full speed ahead, it is expected to create $300 to $400 billion in revenue by 2035, according to a recent McKinsey report.
While many wait on the edge of their seats for driverless cars to become a reality, the market is bustling—not only with buzz and excitement, but also with many unknowns. Today, we’ll address some of the most common questions around fully autonomous vehicles (AVs) to provide a clearer picture of where the industry is headed—and when new technologies are expected to come to fruition.
When will truly autonomous vehicles be available?
As of now, there are no fully self-driving cars available to buy, but several autonomous car test projects are currently navigating city streets. This includes autonomous transport company May Mobility, which is on a mission to make transit more sustainable and accessible by building autonomous vehicle technology and piloting projects in cities such as Grand Rapids, MN, Ann Arbor, MI and Arlington, TX. Alphabet’s Waymo, formerly known as the Google self-driving project, has expanded its public, fully autonomous ride-hailing service to cities such as Los Angeles, CA, and Phoenix, AZ. Santa Monica, CA-based autonomous vehicle technology company Motional has teamed up with rideshare companies such as Uber and Lyft for driverless taxi rides and deliveries. Driverless rideshare company Cruise, based in San Francisco, CA began offering autonomous rides to the public in 2022. There are many other companies currently driving AV initiatives for commercial use.
In the retail market, vehicles that offer advanced driver assistance (ADAS) features that bring them closer to being “self-driving” are becoming more prominent. For example, some vehicles allow hands-free driving on pre-mapped highways, while others provide autonomous driving under a certain speed limit on specific roads.
Research firm GlobalData anticipates that self-driving cars will be available by 2035. Its analyst Amrit Dhami predicts that we’ll see about 5.1 million AVs that can drive themselves within a limited, geofenced area, as well as some 2.7 million AVs equipped to handle all self-driving tasks in all circumstances and environments.
What are the six levels of autonomy?
In autonomous vehicles, there are six levels of autonomy set by the Society of Automotive Engineers (SAE), ranging from level 0 (fully manual) to 5 (fully autonomous). The U.S. Department of Transportation have adopted these levels of autonomy. Here’s the breakdown of the SAE AV classified system:
Level 0: This describes most cars on the road today, including no automation. Vehicles at this level require manual control of the vehicle, and the human performs all the driving tasks. There may be limited systems in place to assist the driver, such as an emergency braking system.
Level 1: This is the lowest level of automation and includes driver assistance. The vehicle features a single automated system, such as monitoring speed through cruise control, while the driver monitors steering and braking.
Level 2: These vehicles have partial automation, also referred to as advanced driver-assistance systems (ADAS). The vehicle can perform steering, accelerating, and decelerating. The human monitors all tasks and can take control at any time. Some vehicles at this level that are currently on the market include Tesla Autopilot and Cadillac (General Motors) Super Cruise.
Level 3: Vehicles at this level have conditional driver automation and “environmental detection” capabilities, which allow them to make informed decisions, like speeding up for a slow-moving car. The driver must remain alert and take control of the car when needed.
Level 4: These vehicles provide high driving automation, which means that the vehicles themselves can intervene if things go wrong or if there is a system failure. However, the human driver can manually override the automation when needed. For now, level 4 vehicles are allowed to operate in self-driving mode within a pre-defined, limited area known as “geofencing,” where speeds reach an average of 30 mph. Most level 4 vehicles available today are geared towards ridesharing and include cars developed by Waymo, as well as French company NAVYA, Canadian automotive supplier Magna, and a partnership between Volvo and Baidu that is being developed for China’s robotaxi market.
Level 5: These vehicles will have full driving automation and will not require any human attention. They won’t have steering wheels or pedals for accelerating or braking and will not have to adhere to geofencing limitations. None of these vehicles are currently available to the general public.
In levels 0 through 3, the human monitors the driving environment. In levels 4 and 5, the automated system monitors the driving environment.
Where do states and countries stand on autonomous vehicle laws?
Some U.S states allow for autonomous vehicles to be used outside of commercial use. These states are Arizona, California, Colorado, Florida, Georgia, Iowa, Nevada, New Hampshire, North Carolina, North Dakota, Oklahoma, Tennessee, Texas, Utah, and West Virginia. Pennsylvania will also join the ranks in July 2023. There are other states that currently only allow for the testing of autonomous vehicles. These states are Connecticut, District of Columbia, Hawaii, Illinois, Maine, Massachusetts, New Mexico, New York, Ohio, Vermont, Virginia, and Washington.
In terms of driverless car patents, the U.S. leads the way, with 135,828. China follows closely behind with 132,844 driverless car patents. Japan has 57,065 and South Korea has 38,097 driverless car patents, according to Select Car Leasing.
Germany is also leading the way for autonomous driving. Long known as a hub for automotive innovation, Germany was the first country in the world to allow autonomous vehicles at level 4 autonomy on public roads without a human driver behind the wheel as backup.
How safe are self-driving cars?
Automakers are continually bolstering safety for autonomous vehicles through technologies such as infrared sensors that detect lane markings, pedestrians, and cyclists in low lighting and certain environmental conditions, radio detection and ranging (RADAR), in which a sensor uses radio waves to determine the distance between obstacles and the sensor, light detection and ranging (LIDAR) that uses a 360-degree sensor that leverages light beams to determine the distance between obstacles and the sensor, among others. Many automated features, from headlight activation and emergency braking to detection and avoidance, have been designed to increase safety.
Many companies have been committed to significant testing for autonomous vehicle safety. Waymo, for example, has logged over 20 million miles on public roads and tens of billions of miles in simulation and has now taken on the task of writing a “framework” for AV safety. Tesla has driven more than three billion miles in Autopilot mode since 2014. While proving the safety of AVs continues to be a work in progress, companies are making headway every day.
What is the difference between ADAS and fully autonomous driving?
It’s important to remember that a vehicle with advanced driver assistance systems (ADAS) features is not synonymous with an autonomous vehicle (AV). ADAS features continue to evolve as they utilize cameras and radar sensors to alert drivers of other vehicles or pedestrians, activate brakes in the event of an emergency, perform adaptive cruise control, and more. However, they rely on a driver to step in when necessary to take control of the vehicle’s functions.
AVs, on the other hand, are designed to manage every part of the trip without the help of a human driver. As a result, an AV requires much more complex hardware, software, and computing power, as well as testing on the road and in the lab to accommodate millions of miles of driving scenarios.
What is the “connected car,” or vehicle-to-everything (V2X)?
The “connected car” refers to connected-vehicle technology that, using internet connectivity and automotive networking, communicates with outside systems. This might include vehicle-to-vehicle communication, along with GPS or apps to unlock or start your car. This is also known as vehicle-to-everything, or V2X, technology, in which vehicles share real-time information with drivers, other vehicles, pedestrians, and roadway infrastructure through sensors, cameras, and wireless connectivity.
What are software-defined vehicles, and how are these different from autonomous vehicles?
A software-defined vehicle (SDV) provides owners with access to safety updates, security, comfort features, over-the-air (OTA) updates, improved vehicle life-cycle management, and more. The vehicle will be updated, enhanced, and will obtain more advanced capabilities throughout its lifetime, much like the way that a consumer can update his or her smartphone. This is a big step ahead of the vehicles of the past that would lose value—and never improve or advance in features—when driven off the dealer’s lot.
A SDV can co-exist with autonomous vehicles. A SDV can be an autonomous vehicle, but it also can be a vehicle that is not fully autonomous and offers ADAS features. With software updates available over the lifetime of a vehicle, a SDV holds the potential to continually evolve and advance with the latest technologies. OEMs envision a future where vehicles will offer technology updates without waiting for new vehicle launches or mid-cycle enhancements. This software can also unleash new insights into driver behaviors, road and environmental conditions, safety, and maintenance, and more—ultimately paving the way (and speeding innovation) for self-driving cars of the future.
Access a collection of resources about driverless cars. Here, you’ll learn more about the intricacies of training an AV’s algorithm, improving safety, bridging the gaps in connected cars, and more at Keysight’s Autonomous Vehicles content hub.