If you’ve seen movies like Blade Runner or I, Robot, then you know that self-driving cars are a sure sign of a sci-fi future.
In the year 2018, the future is closer than ever as major automobile companies like Tesla and General Motors are already developing and fine-tuning the technology for autonomous cars.
Many tech companies are also working hard and investing substantial amounts of money to perfect driverless technology. Fully-driverless tech is still in its testing stages, and there are various technical and ethical challenges that must be overcome before they become commercially available.
Autonomous vehicles have been developing for years, and a guide to levels of autonomy has been adopted by the US Department of Transportation and the United Nations. This guide separates vehicle automation into six different levels ranging from no automation at level 0 to fully autonomous driving at level 5. Each level has a unique set of requirements that an automobile must meet before it is considered to operate at that level.
At Level 0, the autonomous driving capabilities of a vehicle are non-existent. Basic fixed-speed cruise control may be present in these vehicles, but a human is still required to operate it and make every driving decision. Most vehicles manufactured from the early 2000’s and earlier have a level 0 rating for automation.
Most modern passenger automobiles qualify for an autonomous vehicle rating of level 1. In order to meet the requirements, a vehicle must offer at least one advanced driver-assistance feature. The most common assistive features found in level 1 vehicles include adaptive cruise control or automatic braking. At this level, a human driver is still fully in control of driving, but the vehicle is capable of handling one feature for the driver’s convenience. This level of automation is nowhere near enough for a vehicle to be considered a self-driving car.
Level 2 automation takes a step forward towards autonomous driving, but it still has some ways to go. At this level, a vehicle has at least two advanced driver assistance systems (ADAS) that work in coordination with each other to ease the driving experience. However, a driver must actively monitor the vehicle while driving and be prepared to intervene.
Some features that can be considered ADAS include adaptive cruise control, active lane-keeping assistance or emergency automatic braking. The range of sophistication for these features can vary, but they are available on a majority of vehicles produced in 2018. Currently, level 2 is the highest level of autonomous vehicles available commercially.
A vehicle can be considered a driverless car when it qualifies for level 3 automation, at the very least. At level 3, a vehicle is able to take full control of driving when certain operating conditions are met. The autonomous driving is usually limited to driving in the city or on a freeway, excluding exit and entrance ramps. In order to reach this level of automation, a vehicle needs advanced sensor packages, hardware backups and advanced software that prioritize occupant safety.
Drivers in a level 3 vehicle must continue to monitor the road, even when the vehicle is self-driving. In the event of an emergency, the driver must retake control of the vehicle and react appropriately. The google self-driving car reached level 3 autonomy in 2012, but there were some problems that arose during testing. Google found that the human drivers were too slow to retake control of the vehicle in a crisis, most likely because they were too trusting of the car’s capabilities. Some companies are still trying to make level 3 autonomous cars available to the public, but many countries have yet to legally approve of them.
Once vehicles reach level 4 of automation, they truly become autonomous vehicles. At this level, an automobile can complete entire journeys without a driver’s involvement. Nevertheless, a level 4 autonomous car still has limitations such as the guidance system being confined to familiar territory and consistent conditions. A level 4 automated vehicle is expected to have a steering wheel and pedals just in case a human driver needs to assume control.
The end goal of all self-driving car developers is to reach full automation at level 5. A level 5 car can drive itself wherever it needs to go and navigate using systems that include a combination of advanced hardware and sophisticated software. These vehicles are expected to have no steering wheel or pedals for human control. Unfortunately, automobile and tech industry experts all agree that they are quite far from reaching a commercially viable level 5 vehicle.
Nowadays, there are hundreds of self-driving cars driving around America. Hundreds of vehicles are currently being tested in a handful of states. Google has its Waymo self-driving car being tested in Texas, Georgia, Michigan, California, Washington and Arizona. Since these cars learn and adapt by spending time on the road, drive-testing them is vital for their advancement. Currently, automobile and tech companies have over 10 million miles of driving experience for their test vehicles.
Unfortunately, the testing hasn’t gone without complications. While autonomous vehicles are safer overall, crashes can still occur due to both human error and technical fault. There have been a handful of accidents and a few ill-fated fatalities involving self-driving cars, which has caused some states to tighten restrictions on testing. Still, experts believe that these autonomous automobiles can reduce traffic incidents caused by human factors like distracted and drunk driving.
Although there have been a few roadblocks along the way, autonomous cars still have a hopeful future. Testing from companies like Uber and Google continues as they try to perfect the technology. Many states are competing to become future testing locations for self-driving cars. States that allow testing can benefit financially through major corporate partnerships and potentially safer roads in the future.
On the other hand, there are still legislative barriers in many states. Some states have stricter rules and limitations on autonomous vehicles than others. The lack of a national standard on these vehicles makes them difficult to test across the country. Safety concerns, which include accidents and potential hacks, are a major reason why these barriers stand in the way of driverless cars.