The 6 Autonomous Driving Levels Explained
In this post, we will explore the 6 levels of autonomous driving and discuss the current industry standards and potential future progress of autonomous vehicles.
Created on September 3|Last edited on December 28
Comment
There are six autonomous driving levels, classified based on their degree of automation. Categorized by the Society of Automotive Engineers (SAE), the spectrum ranges from level 0 to level–from no automation to fully autonomous vehicles.
At a high level, you can think of the levels of automation as:
- Level 0: No automation at all
- Level 1: Very light automation (cruise control, etc.)
- Level 2: Some automation but requires human attention at all times
- Level 3: Can self-drive but require intervention in severe conditions
- Level 4: Highly autonomous
- Level 5: Completely autonomous
In essence, autonomous vehicles at levels 0, 1, and 2 require human intervention at all times while levels 3, 4, and 5 are “self-driving.”
6 Levels of Autonomous Cars, Image by Author
Table Of Contents
Table Of ContentsThe State Of Autonomous VehiclesThe Levels Of Autonomous DrivingAutonomous Driving Level 0: No Driving AutomationAutonomous Driving Level 1: Driver AssistanceAutonomous Driving Level 2: Partial Driving AutomationAutonomous Driving Level 3: Conditional Driving AutomationAutonomous Driving Level 4: High Driving AutomationAutonomous Driving Level 5: Full Driving AutomationThe Future Of Autonomous VehiclesRecommended Reading
The State Of Autonomous Vehicles
Autonomous vehicles are an emerging field of technology with heavy investments made by myriad billion-dollar corporations. One of the major benefits of autonomous vehicles that makes them so attractive is their ability to reduce human errors during driving and increase life satisfaction for humans: a reduction of boring commutes, unproductive time spent in traffic, and accidents.
Object Detection for Autonomous Vehicles (A Step-by-Step Guide)
Digging into object detection and perception for autonomous vehicles using YOLOv5 and Weights & Biases
The nuScenes Dataset
nuScenes is a large-scale 3D perception dataset for Autonomous Driving provided by motional. The dataset has 3D bounding boxes for 1000 scenes.
The Levels Of Autonomous Driving
Let's dig into the six levels of autonomous vehicles in more depth and explore the current industry standards–and our potential future progress.
For a more clear understanding of the concepts, we're going to categorize each level into three main factors: Human requirements, vehicle features, and examples.
Autonomous Driving Level 0: No Driving Automation
At this level, there is no automation in vehicles. Almost all cars on the road today belong to this category. They might have some features like cameras that assist when you're backing up or collision warning systems but they have no control over the movement of the vehicle.
Level 0 - No Driving Automation, Image by Author
Level 0 Summary
- Human requirements: Humans do all the work, they are always in control of the vehicle
- The vehicle features: Automatic Emergency Braking, Lane Departure Warning, Blindspot Warning, Back Cam, Collision Warning, etc.
- Example: Almost every car out there
Autonomous Driving Level 1: Driver Assistance
Level 1 is the next level on the automation spectrum. Here, the vehicle has the capacity to perform one and only one autonomous task at any given time. These tasks are related to safety and provide basic movement assistance like steering or braking.
Level 1 - Driver Assistance, Image by Author
If it provides both steering and braking, then it belongs to Level 2.
💡
Level 1 Summary
- Human requirements: Humans do most of the work and are in control of the vehicle at all times.
- Vehicle Features: Lane centering, adaptive cruise control
- Example: 2020 BMW Series 3, 2020 Volvo S60
Autonomous Driving Level 2: Partial Driving Automation
Level 2 is the third level in the automation spectrum.
At this level, all the vehicles possess some form of Advanced Driving Assistance Systems (ADAS). ADAS will take control over steering, braking, and acceleration in very specific situations, including low-traffic environments like highways.
Level 2 - Partial Driving Automation, Image by Author
Level 2 Summary
- Human requirements: the driver must remain alert throughout and is required to intervene in many situations
- Vehicle features: Lane centering and adaptive cruise control simultaneously
- Examples: Highway driving assistants used in the 2021 Hyundai Palisade and Teslas with Autopilot
Autonomous Driving Level 3: Conditional Driving Automation
Level 3 is the fourth level in the automation spectrum. It is a significant technological improvement over level 2 as the vehicle is mostly self-driven, though it introduces new issues and we'll read below.
Level 3 autonomous vehicles run mainly on their own and require human intervention only in cases of extreme environments and failures.
Attaining this level of autonomy introduces a lot of usability challenges. Consider a situation in which the car passes control of the vehicle over to the human driver in a very extreme situation (e.g., during an accident). This is one of the reasons for the jump of development from level 2 to level 4, altogether skipping this level.
Level 3 - Conditional Driving Automation, Image by Author
Level 3 Summary
- Human requirements: The driver must be alert at all times during the drive. The driver should take control of the car during extreme conditions.
- Vehicle features: Traffic Jam Chauffeur.
- Example: Mercedes Benz is one of the leading manufacturers in the development of level 3 cars. Their Drive Pilot System can navigate traffic and detect weather conditions.
Autonomous Driving Level 4: High Driving Automation
This is the fifth level in the automation spectrum. At level 4 autonomy, there is no need for human intervention in driving. The system drives on its own, and we are tagging along for the ride, potentially eliminating the need for steering wheels and pedals.
One major limiting factor to level 4 autonomous vehicles is the use of geofencing. These vehicles are geofenced to particular areas and won’t be able to drive anywhere other than those places. Severe weather conditions would also affect these vehicles and would likely disturb their operation.
Level 4 - High Driving Automation, Image by Author
Level 4 Summary
- Human requirements: Enter your destination and have some fun (maybe read a book, or put in some time on an installed rowing machine).
- Vehicle features: Driverless Taxis and Transportation Systems.
- Example: Waymo, Google’s autonomous vehicles products are targeted to work at this level. As of writing, it is operating only in Phoenix.
Autonomous Driving Level 5: Full Driving Automation
Level 5 is the highest level in the autonomous driving spectrum. These vehicles can drive themselves in all conditions. They are not bounded by geofences and can travel anywhere. They can safely transport humans in severe weather and on damaged roads.
Currently, there are no real examples of this level outside of science fiction (say, the iconic Batmobile). Using GPS, the Batmobile can track Batman and reach him when the need arises. It can also easily move through traffic, damaged roads, and severe weather.
Level 5 - Full Driving Automation, Image by Author
Level 5 Summary
- Human Work: Enter the destination and relax, read books or sleep.
- Vehicle Features: Driverless transportation with the ability to go anywhere.
- Example: Batmobile and Lexus 2054.
The Future Of Autonomous Vehicles
As of autumn 2022, we are currently able to produce level 3 autonomous vehicles but they are not allowed on the roads in the US (though they are in Germany).
Some experts are suggesting that level 4 will be available in 2024 or 2025, though mass adoption will likely lag behind, due to costs and of course, legislation.
As touched on above, early adoption of level 4 will likely begin in the taxi and transport industries where their ROI is highest. This will likely assist in producing the scale to drive down costs, making level 4 (and then hopefully level 5) available to all.
Recommended Reading
The Many Datasets of Autonomous Driving
Below we'll explore the datasets used to train autonomous driving systems to perform the various tasks required of them.
Autonomous Vehicle Companies and Their ML
The race in the autonomous vehicle space has resulted in a large number of companies vying for a position at the top. Here's what some of the main ones are doing.
The ML Tasks Of Autonomous Vehicle Development
This report goes through the different tasks in the autonomous vehicle development lifecycle and the various machine learning techniques associated with them.
A System of Record for Autonomous Driving Machine Learning Models
A look at the most useful Weights & Biases features for autonomous vehicle companies
Add a comment
Tags: Autonomous Vehicles, Beginner
Iterate on AI agents and models faster. Try Weights & Biases today.