autonomous vehicles
This article is also available here in Spanish.

These are the 6 levels of autonomous driving that explain when cars will drive themselves

My list

Author | Jaime Ramos

Cars don’t drive themselves, yet. However, we are currently in the midst of a historic period in which we are closer to the reality of self-driving cars. The six levels of autonomous driving help us understand how far we have come, how far we still have to go and, above all, how to distinguish the various categories of driving assistance.

How do autonomous vehicles work?

The autonomous vehicle does not involve developing just one technology. It is the sum of innovations, each with their own challenges. Reaching the ultimate level also involves reaching maturity in terms of sensor technology, the deployment of communication infrastructures between the vehicle and the road, or in the application of Big Data and AI.

The autonomous driving levels explain how that narrative should be in terms of its challenges, but also with virtues from which we are already benefiting. This is the case of the ADAS advanced driver assistance systems that prevent deaths, in the hope that full autonomous driving keeps its promise of reducing accident rates to anecdotal levels. According to the EU, if 30% of vehicles were equipped with advanced assistance technologies, accidents would drop by 15%.

The 6 levels of autonomous driving

The 6 levels of autonomous driving are based on the J3016 scale designed by the SAE. The industry has been using this official classification system since 2014, which unifies criteria around its six stages.

Level 0: No driving automation

The vehicle possesses no automated driving features or ADAS assistance. The driving experience is entirely in the hands of the driver.

Level 1: Driver assistance

The lowest level of automation, the system incudes assistance systems seen in existing models: cruise control, assisted parking or lane keeping assistance. Some also add emergency braking systems. In some parts of the world vehicles include these features as standard, such as in the EU, where they are mandatory.

Level 2: Partial driving automation

In level 2 of autonomous driving, the intervention of the vehicle is intensified, performing various tasks at the same time.

The driver still has to hold the steering wheel, which is not the case in some models that incorporate "automatic pilots" (some Tesla models). Entities such as Euro NCAP warned of how dangerously confusing the hype is.

Level 3: Conditional driving automation

autonomous driving 2

This is a mature automation level, but supervised at all times by the driver, who can and should take control when necessary. This level is now a reality in the car sector. The Volvo EX90, for example, now includes the sensors and electronic management that this level requires.

Level 4: High driving automation

In this phase, if things go wrong, the vehicle does not require human interaction in some circumstances. It is capable of interacting in these circumstances totally by itself. If it requires human intervention, the vehicle will let the driver know and guarantee safety until the human takes control.

Level 5: Full driving automation

The last level is that in which the vehicle does not rely on any external or human factor to move. It is capable of driving by itself. The last two levels are still in the experimental phase. There is no consensus among experts regarding whether or not human intervention should be allowed or if elements such as the steering wheel or the pedals should be eliminated.

However, it will not be long before these questions are resolved. In the coming years, the sector will continue to advance through the six levels of autonomous driving.

Images | Flickr/Daniel Ramirez, Volvo

Related Content

Recommended profiles for you

MG
Milagros granizo
mtspain
ceo
CA
Cecilio Angulo
Universitat Politècnica de Catalunya - AI Research Centre IDEAI
Head Director. Research Centre on Intelligent Data Science and Artificial Intelligence IDEAI-UPC
PU
Pertz Uwe
Swarco Traffic Systems GmbH
Head of Parking and E-Mobility Solutions
LA
luis alcazar
Tukuy Innovation SAC
Consultant
MG
Marilda Gougou
MAS SA
MB
Mahtab Baghaiepoor
Smart City Commission at Yazd City Council
Consultant
RJ
Ryan Jin
EcoLabs Centre of Innovation for Energy, NTU Singapore
Lead (Smart Cities & Future Mobility)
ML
Mladen Lovrenčić
Dyvolve d.o.o.
Managing director
CC
Carlo Conti
ACI
Director of Group Strategy
MP
Maria Paschalidou
OTE SA
ICT Senior Sales Specialist
AN
Alessandros Nunes
VALID
As an IoE (Internet of Everything) architect, I develop new business and technology solutions .
SB
Stefan Bergstrom
City of Sundbyberg
Deputy mayor for urban planning in Sundbyberg, the most densely populated city in Sweden.
KG
Kevin Gitau
Urban design and planning
Urban Designer
CB
Cornel Barbut
Vegacomp Consulting
CEO
AD
Aryo Dhaneswara
IAAC
Arhictect
EB
Erick Bacchin
Esporte Clube Pinheiros
Project Management Special
MH
Mehdi Hasanzadeh
Rahyaft consulting engineers
CEO
YM
Yordan Mantilla
INSE GROUP SAS
CEO / Inse Group SAS
RG
Rodolfo Guija
Guitraffic
CEO
SB
Sadjad Bazarnovi
Smart Tehran Center
Smart Tehran Center Consultant