Written by Grant Malloy Smith, the data acquisition expert
In this series of articles, you will learn about ADAS - Advanced Driver Assistance Systems. We will cover the topic in enough depth that you will:
- Understand what ADAS is and its purpose
- Learn about the sensors and technologies behind ADAS
- See how ADAS systems are tested
Jump to a section
What’s the Purpose of ADAS?
ADAS (Advanced Driver Assistance Systems) are passive and active safety systems designed to remove the human error component when operating vehicles of many types. ADAS systems use advanced technologies to assist the driver during driving, and thereby improve drivers’ performance. ADAS uses a combination of sensor technologies to perceive the world around the vehicle, and then either provide information to the driver or takes action when necessary.
ADAS systems are being applied today to cars, trucks, and buses, as well as farming, construction, and military vehicles.
According to the NHTSA (the National Highway Traffic Safety Administration), more than 36,000 Americans were killed in automobile accidents in 2019 alone. The NHTSA’s August 2016 Traffic Safety Facts Research Note reported that 94% of those accidents were caused by human error, i.e., mistakes made by the driver.
Given this reality, it is easy to imagine how many lives could be saved by effective ADAS systems that prevented many of these errors. In fact, the IIHS (the Insurance Institute for Highway Safety), estimates that even the ADAS technologies available right now could prevent or lessen the effects of 1.8 million accidents every year, and potentially save up to 10,000 lives per year.
There are several distinct levels of ADAS, from simple backup cameras and blind-spot warning sensors to lane departure warning systems, adaptive cruise control, self-parking, and more. The ultimate extension of ADAS will someday be vehicles that are truly autonomous, aka “self-driving,” and which will not require a human driver. But even now, years before we reach fully autonomous cars, ADAS safety features are making driving safer and safer every day.
Companies all around the world are making billions of dollars in investments in developing ADAS technologies. Today it is hard to find an automobile manufacturer that is not budgeting significant resources on ADAS technology. General Motors, Volvo, Toyota, Ford, Volkswagen, Tesla, BMW, Audi, to name just a few. It's a long list made up of virtually every car, truck, and bus manufacturer in the world.
ADAS technology is evolving rapidly, and no one wants to be left behind.
How Does ADAS Work?
ADAS works by alerting the driver to danger or even taking action to avoid an accident. ADAS-equipped vehicles can sense the environment around themselves, process this information quickly and accurately in a computer system, and provide the correct output to the driver.
ADAS-equipped vehicles have an array of advanced sensors that augment the eyes, ears, and decision-making of the human driver. Can you see in the dark? Not very well, but RADAR can. Can you echolocate like a bat or a dolphin to determine if there’s a child behind your car before you put your car in reverse? No, but SONAR sensors can. Can you see in all directions at once? No, but cameras and LiDAR sensors can. Do you know your exact latitude and longitude at all times? No, but several constellations of global positioning satellites in space can send that information to your car, and more.
Different types of ADAS sensors used in today's autonomous vehicle
The ADAS system architecture consists of a suite of sensors, interfaces, and a powerful computer processor that integrates all of the data and makes decisions in real-time. These sensors are constantly examining the environment around the vehicle and providing this information to onboard ADAS computers for prioritization and action. Today, they are saving lives by preventing accidents that would have happened without ADAS. Someday, these technologies will lead to fully autonomous vehicles.
Active vs. Passive ADAS Systems
Passive ADAS Systems
Regardless of the number or types of sensors installed, in a PASSIVE ADAS system, the computer merely informs the driver of an unsafe condition. The driver must take action to prevent that condition from resulting in an accident. Typical warning methods include sounds and flashing lights, and sometimes even physical feedback, for example, a steering wheel that shakes to warn the driver that the lane they are moving into is occupied by another vehicle (blind spot detection).
Common Passive ADAS Functions Include:
- ABS - Anti-lock Braking Systems: Keep the car from skidding and turning when emergency braking is applied.
- ESC - Electronic Stability Control: Assists the driver in avoiding under or over-steering, especially during unexpected driving conditions.
- TCS - Traction Control System: Incorporates aspects of both ABS and ESC above, to assist the driver in maintaining adequate traction when negotiating turns and curves.
- Back-up Camera: Provides the driver a view behind the car, when parking or backing up.
- LDW - Lane Departure Warning: Alerts the driver if the vehicle is not keeping within its lane.
- FCW - Forward Collision Warning: Tells the driver to brake in order to avoid a collision ahead.
- Blind Spot Detection: Warns the driver that there is a vehicle in their blind spot.
- Parking Assistance: Warns the driver when their front or rear bumpers are approaching an object at low speeds, i.e. when maneuvering into a parking space.
Self-driving test vehicle with sensor array on the roof.
Alena Nesterova, CC BY-SA 4.0, via Wikimedia Commons
Active ADAS Systems
In an ACTIVE ADAS system, the vehicle takes direct action. Examples of Active ADAS functions include:
- Automatic Emergency Braking: Automatically brakes as required to avoid hitting a vehicle ahead or another object, including pedestrians, animals, or anything in the lane of travel.
- Emergency Steering: steers the car to avoid striking an object in the lane of travel.
- Adaptive Cruise Control: Adjusting cruise control speed to match vehicles ahead.
- Lane Keeping Assist and Lane Centering: Steering the car to stay centered in the lane.
- Traffic Jam Assist: Combines adaptive cruise control and Lane Keeping Assist to provide semi-automated driver help during dense traffic events, i.e., stop and go conditions due to lane closures, road construction, etc.
- Self Parking: Self-maneuvering into parking spaces.
Transport Canada has put together an excellent overview of both passive and active ADAS functionality:
Warns drivers of a vehicle in their blind spot.
|Forward collision warning
Detects and warns the driver of a potential collision with a vehicle ahead. Some systems include pedestrian or other object detection.
|Lane departure warning
Monitors the vehicle's position within the driving lane and alerts the driver as the vehicle is drifting over the lane markings.
|Parking collision warning
Detects obstructions near the vehicle during parking maneuvers.
|Rear cross-traffic warning
Detects vehicles approaching from the side and rear of the vehicle while traveling in reverse and alerts the driver.
|Automatic emergency braking
Detects a potential collision with obstacles ahead, provides forward collision warning and automatically applies the brakes to avoid or lessen the severity of the impact. Some systems include pedestrian or other object detection.
|Automatic emergency steering
Detects a potential collision and automatically controls steering to avoid or lessen the severity of the impact. Some systems include pedestrian or other object detection.
|Reverse automatic braking
Detects a potential collision while traveling in reverse and automatically applies the brakes to avoid or lessen the severity of the impact. Some systems include pedestrian or other object detection.
|Driving Control Assistance|
|Adaptive cruise control
Assists with acceleration and/or braking to maintain a prescribed distance between it and a vehicle in front. Some systems can come to a stop and continue.
|Active driving assistance
Assists with vehicle acceleration, braking, and steering. Some systems are limited to specific driving conditions.
Assists with the steering to maintain the vehicle within the driving lane.
|Active parking assistance
Controls steering and potentially other functions like braking and accelerating during parking. The driver may be responsible for the acceleration, braking, and gear position. Some systems are capable of parallel and/or perpendicular parking.
|Remote parking assistance
Parks vehicle without the driver being physically present inside the vehicle. Automatically controls acceleration, braking, steering, and shifting.
Assists the driver with visual guidance while backing towards a trailer or during backing maneuvers with a trailer attached. Some systems may provide additional images while driving or backing with a trailer. Some systems may provide steering assistance during backing maneuvers.
|Other Driver Assistance Systems|
|Advanced forward lighting systems
Prompts lights to automatically adapt to changing driving conditions by swiveling to illuminate the vehicle's travel path, switching from high beam to low beam, or shining light 90 degrees in either direction at an intersection.
Provides a view of the area behind the vehicle when in reverse. Could include trailer assistance, a system that assists drivers during backing maneuvers with a trailer attached.
Brake assist monitors brake pedal pressure to automatically sense emergency braking. It then boosts brake pressure to levels beyond those of the driver's pedal and more quickly to shorten stopping distances.
Monitors drivers to determine if they are actively engaged in the task of driving. Some systems monitor the driver’s eye movement and head position.
|Electronic stability control
Automatically brakes one or more wheels for short periods of time and/or reduced engine power to keep the vehicle moving in the intended direction when it swerves to avoid an obstacle.
Projects an image of vehicle data and/or navigational information into the driver's forward line of sight.
Aids driver vision at night by projecting enhanced images on the instrument cluster or head-up display.
|Roll stability control
Limits vehicle roll by braking one or more wheels and reducing engine power in extreme cornering or evasive maneuvers.
Reminds drivers of the current speed and/or alerts drivers when they drive above the speed limit.
Uses cameras located around vehicles to present a view of surroundings.
|Tire pressure monitors
Monitors the air pressure of all the wheels and alerts the driver when a tire's pressure has dropped below a safe level.
Monitors wheel speed and limits wheel spin when accelerating by braking and/or reducing engine power to the drive wheels.
Types of Driver Assistance Technologies, courtesy of Transport Canada
The ultimate extension of ADAS technology is full autonomy, i.e., self-driving capability (a common acronym is AV - “Autonomous Vehicle”). If we imagine a future world where cars, trucks, and buses are driving around without human operators in control, the challenges involved in creating such a broad system of automated driving technologies seem daunting.
But the rewards will be great:
- Far fewer accidents caused by driver error, meaning fewer human injuries and death, as well as fewer damaged or destroyed vehicles and property.
- Lower energy consumption due to ride-sharing, and possibly even fewer cars needed per capita.
- Self-managed traffic systems on busy roads, meaning fewer traffic jams.
- And many others.
Highways near urban areas are prime candidates for traffic metering
In 1963 Chicago introduced “ramp meters” (traffic lights) at the on-ramps to busy highways, to regulate the number of cars that could enter during busy times of the day and prevent traffic jams. This practice is found all around the world today. In the future, this kind of regulation could be done at a higher level, pacing or even redirecting cars to prevent traffic jams.
The Six Levels of Vehicle Autonomy
The US Department of Transportation has adopted the six levels of vehicle autonomy developed by the SAE (The Society of Automotive Engineers International), as follows:
|0||No Autonomy||The driver is 100% in control of the vehicle. There can be systems like anti-lock brakes installed, but they do not “drive” the vehicle.|
|1||Driver Assistance||The lowest level of automation, where a single system such as passive cruise control or adaptive cruise control, is present to assist the driver.|
|2||Partial Driving Automation||Level 2 vehicles have an on-board ADAS system that can steer, accelerate, and brake without human intervention. However, a human must be in the driver’s seat and able to take over at any time, hence the “partial” in the title.|
|3||Conditional Driving Automation||A significant step up from Level 2, these vehicles can make decisions based on traffic and other considerations, and then act on them. A human operator is still required to be in the driver’s seat and able to take over at any time.|
|4||High Driving Automation||These vehicles are self-driving, but at this point in time, they are limited to operate only within certain geographies, roadways, or within certain speed limits. Taxi and ride-sharing services in several markets are deploying level 4 vehicles today.|
|5||Full Driving Automation||Not yet available to the general public, a level 5 car does not need any human interaction at all. In fact, a level 5 car or truck would not need to have a steering wheel or brakes, or any provision for a human driver.|
Types of Autonomous Vehicles
There are several types of autonomous vehicles today:
Completely self-driving passenger cars and taxis remain the ultimate goal of this technology. But like America’s moon missions of the 1960s and 1970s, a huge array of benefits, both inside and outside of space exploration, resulted from the technical challenges that had to be overcome to send men to the moon and return them safely to Earth.
So today, years before 100% autonomy has been reached, we are already reaping the benefits of advancements brought about by ADAS technology. With functions like collision avoidance, backup cameras, parking sensors, adaptive cruise control, and blind-spot detection, ADAS systems have already made driving safer.
Tesla’s level 2 “Autopilot” is not the most advanced self-driving technology on the market, but it is probably the best known by consumers because of the popularity of the Tesla brand of cars. Tesla gained fame as a maker of completely electric passenger cars. They are expanding into the small (pickup truck) and large (Class 8) truck space in 2021 and 2022.
Self-driving and electric car technology fit well together since both are very computer-intensive. In July 2021, Tesla began downloading the long-awaited Beta 9 version of level 4 autonomy into their customer’s cars. At the same time, Tesla is quick to caution that drivers must remain in the driver’s seat and be able to take over at any moment. Tesla CEO Elon Musk began promising this beta version as far back as 2018.
Yandex has been testing its self-driving vehicles in various cities around the world. Have a look at this time-compressed video of a one hour drive through the streets of Ann Arbor by one of their cars, as it navigates construction areas, pedestrians, and other vehicles:
Yandex 1-hour autonomous drive in Ann Arbor, Michigan
Special provisions are needed by authorities in these towns and cities to allow test and commercial deployment of self-driving vehicles. There are risks, of course, and new laws will need to be written to deal with liability, for example, when accidents inevitably occur.
Taxis and Ride-Sharing Vehicles
Companies working on self-driving taxis (aka “robo-taxis”) and shared-ride spaces include Waymo, TuSimple, Plus AI, and more. Not surprisingly, ride-share giants, Uber and Lyft are deeply involved in the self-driving space, investing millions and forming alliances across the self-driving landscape. They are looking to replace the cost of human drivers with technology, just as the large and small truck-based delivery companies are.
Typical Waymo One driverless taxi
Image courtesy of Waymo, Inc. © 2021 Waymo LLC
Looking at one example, Waymo is an American subsidiary of Alphabet, the parent company of Google. Today, people in the metropolitan Phoenix area in the USA can hail a fully autonomous Waymo One taxi using the Waymo app on their smartphones. Waymo driverless taxis are considered to be at level 4 of autonomy, but only within a prescribed route and under known conditions.
As of this writing, Waymo is operating 300 taxis in a ~100 square mile area (~260 km2) that includes the towns of Chandler, Gilbert, Mesa, and Tempe. It should be mentioned that fully driverless, aka “rider only” Waymo taxis operate in an area about half that size.
Transportation Trucks, Large and Small
According to the US Department of Energy (US DOE), each year vehicles transport 11 billion tons (9.9 B metric tons) of freight - roughly $35 B (€29.5 B) worth of goods each day, and move people more than 3 trillion miles (~4.8 T km). The transportation sector accounts for approximately 30% of total US energy needs, and 70% of US petroleum consumption.
The scale of these numbers is why the DOE is so interested in increasing the energy efficiency of transportation, and advanced ADAS features are an essential component. How?
One of the earliest applications envisioned for self-driving or autonomous vehicles was in the area of “platooning” trucks. In this scenario, first put forward by Popular Science magazine in 1995, a lead truck driven by a human operator would lead a convoy (aka “platoon”) of autonomous trucks that would follow a human-operated leader. This approach has applications not just in transportation, but in farming, mining, military operations, and more. Backed by Volvo, Peloton Technology is actively developing truck “platooning” technology.
According to a 2018 DOE report: “Platooning involves the use of vehicle-to-vehicle communications and sensors, such as cameras and radar, to virtually connect two or more trucks together in a convoy. The virtual link enables all of the vehicles in the platoon to communicate with each other, allowing them to automatically accelerate together, brake together, and enables them to follow each other at a closer distance than is typically possible with unlinked trucks.”
Advantages of platooning include:
- Reduced energy costs, because the lead vehicle cuts through the air, reducing the aerodynamic drag on platooned trucks following very closely behind.
- Because there are smaller distances between them, platooned trucks occupy less space on the roads.
- Increased human safety, since there is only one driver. This is even more significant in the platooning of military vehicles operating in war zones.
Today the emphasis on self-driving commercial vehicles is for freight delivery in large trucks, as well as local delivery of food and small packages in much smaller trucks, sometimes referred to as “last mile” delivery. This includes packages as well as food, like pizza delivery, for example. Numerous companies are in the test and small-scale operational phases of development in this area. Companies working in this area include Daimler, Volvo, Navistar, Paccar (the maker of Peterbilt, DAF, and Kenworth trucks), Ford Otosan, and more.
Waymo Via is Waymo’s commercial delivery service, applying the same driverless vehicle technology from their taxi service to both long-haul trucking as well as last-mile delivery using much smaller vehicles. Waymo does not manufacture vehicles but has partnered with other companies like Fiat-Chrysler, Audi, Toyota, and Jaguar.
Tesla is developing the Tesla Semi, a class 8 semi for long haul trucking. It will feature four independent electric motors - one per drive wheel. The first version will require a driver, but it is intended to become a completely autonomous truck based on the company’s Autopilot system.
The upcoming Tesla Semi, all-electric truck
Image courtesy of Tesla, © 2021 Tesla
When you were a child, did you ever think your family car would be outfitted with RADAR and SONAR as airplanes and submarines had? Did you even know what LiDAR was? Did you imagine flat-screen displays dominating the dashboard and a navigation system connected to satellites in space? It would have seemed like science fiction, and utterly out of reach for 100 years at least. But today, all of that and more are a reality.
ADAS is the single-most-important type of development going on today. Of course, there are hybrid and electric-powered developments going in parallel, which are also extremely important for reducing greenhouse gases and the use of fossil fuels.
ADAS goes directly to the most important aspect of travel: human safety. Since more than 90% of road accidents, injuries and fatalities are due to human error, every advancement in ADAS has a clear and absolute effect on preventing injuries and deaths.