Autonomous The 5 levels of autonomous driving explained
Autonomous driving will change the way we travel. Reduced traffic congestion, lower travel costs, and no more circling for parking spaces will make our daily commutes quicker, less stressful, and more affordable. It'll also reduce harmful CO2 emissions, improving the quality of air that we breathe.
Fully-automated vehicles may not be the norm just yet, but the reality isn't as far away as we might think. There are five levels of driving automation that specify how capable a vehicle is of operating on its own, without human intervention. The classification system was set by SAE International (Society of Automotive Engineers) in 2014 and is often used as a reference point in discussions surrounding vehicle automation.
Level 0 autonomy refers to a typical, everyday car. The driver performs all operations, including steering, accelerating, and braking and the vehicle has no autonomous or self-driving controls at all. Practically every road vehicle offers level 0 autonomy. A typical example would be the 2018 Honda Jazz.
Let's look at the five levels in more detail:
Level 1: driver assistance
At this level, the driver still handles most of the car's functions but with a little autonomous help. For example, a level one vehicle might provide you with a brake boost if you edge too close to another vehicle, or it might have an adaptive cruise control function to control your distance and speed.
Likewise, Level 1 autonomous vehicles might have a park assist function, where a beeping sound alerts the driver to an approaching obstacle. Level 1 autonomy is common in most cars today, and a typical example would be the 2018 Nissan Sentra, with its Intelligent Cruise Control feature.
Level 2: partial automation
Partial automation enables drivers to disengage from some driving functions. Level 2 vehicles are able to assist with functions like steering, acceleration, braking, and maintaining speed, although drivers still need to have both hands on the wheel and be ready to take control if necessary.
With steering, Level 2 vehicles assist by centering the car within the lane, whereas the speed control function ensures that the correct distance is kept from other cars. An example of a car with Level 2 autonomy is the 2019 Volvo S60, with its auto-braking feature and Pilot Assist capabilities.
Level 3: conditional automation
Conditional automation allows drivers to sit back and let the car do all the driving. Also referred to as 'eyes-off' vehicles, drivers are able to focus their attention on other activities like using a mobile phone, for example. Many Level 3 cars don't require any human intervention at all when driven at a speed of less than 60 km/h.
At this level, cars can be considered truly autonomous, but only under ideal road conditions. There aren't many (if any) level three vehicles driving on public roads that aren't limited-access highways. However, Honda is reportedly set to introduce a level three car on public freeways this year.
Level 4: high automation
At Level 4, vehicles are capable of steering, accelerating, and braking on their own. They're also able to monitor road conditions and respond to obstacles, determining when to turn and when to change lanes. Level 4 autonomous driving can only be activated when road conditions are ideal.
At this level, vehicles can't negotiate more dynamic conditions like traffic jams or other major obstacles. The best example of a Level 4 autonomous vehicle is Google's Waymo project in the U.S.
Level 5: full automation
Level 5 autonomous driving requires no human interaction. Vehicles are able to steer, accelerate, brake and monitor road conditions like traffic jams. Essentially, Level 5 automation enables the driver to sit back and relax without having to pay any attention to the car’s functions whatsoever.
Vehicles will be driven using Artificial Intelligence (AI) and will respond to real-world data points, generated from sensors. In a previous article about AI and mobility, we highlighted that a huge amount of data is produced in autonomous vehicles—as much as 4TB per hour. Only a powerful computing system like Artificial Intelligence can process such large volumes of data quick enough to achieve real-time responses.
Although we’ve yet to see a vehicle reach true Level 5 autonomy, Audi’s AI:CON design study provides a glimpse into what a fully-automated car might look like.