Advanced Driver Assistance System (ADAS)
What is ADAS?
Vehicle accidents are usually caused by human error, so to reduce or prevent these entirely, an increasing number of vehicles are fitted with advanced driver assistance systems, commonly known as ADAS. These systems use multiple real-time vision-based and sensor-based algorithms fed from a variety of sensors, such as cameras, to support passive real-time informational, and active decision-making systems, which are designed to reduce accidents and save lives.
ADAS features include pedestrian detection, lane departure warnings, traffic-sign recognition, automatic emergency braking, and blind-spot detection. These ADAS applications are considered to be Level 2 or Level 2+ systems on the six SAE levels of autonomous driving. This requires the driver to be “in the loop”; that is, in full control of the vehicle at all times.
Common examples of ADAS
Rather, than a full, comprehensive list of all ADAS features, here we will highlight a selection of important examples of ADAS that are commonly used in today’s world.
The most familiar form of ADAS are the alerts and warnings from a vehicle which are triggered by certain actions. Parking sensors are a common example. These scan the area around the car at low speeds and inform the driver via audio warnings that become louder and faster as the driver gets closer to an object.
Another common example is tyre pressure warnings, which notify the driver via a dashboard light when tyre pressures drop below a certain value.
Using newer technology, and therefore a less common example of ADAS is driver drowsiness detection. With this, the vehicle uses AI to obtain information such as facial patterns, steering movements, pedal actuation, velocity, and turn-signal use to decide whether the driver is drowsy. If it believes they are, it will then trigger a loud alert or vibrate the driver’s seat, signalling them to pull over, to avoid falling asleep at the wheel.
There are many uses for ADAS. An example of an ADAS that is commonly built into modern vehicles is an anti-lock braking system (ABS). First made available in the 1970s it uses sensors used to detect wheel speed and prevent the wheels from locking under heavy braking to prevent skidding. Brake Assist is a system that detects when the driver is attempting an emergency brake and assists with full braking pressure to ensure the car stops far sooner. Automated Emergency Braking uses sensors to detect a potential crash and apply the brakes automatically.
Rain sensors trigger automatic actions such as activating windscreen wipers at the appropriate speed for the rainfall. In newer cars, they are also used to automatically close open windows and convertible car hoods.
Automated parking is something we are likely to see more of as autonomous driving becomes more prevalent. This would be where the vehicle takes full control of the parking functions—including steering, braking, and acceleration—while monitoring the surroundings and obstacles that may be in the way.
For many ADAS features to work, the vehicle must have its own visual and environmental monitoring system, either based on cameras, radar, lidar or both. With these, the vehicle can assist the driver with things such as traffic sign recognition, pedestrian detection in low visibility, night vision, as well as a full 360-degree view of the vehicle itself.
As these advancements increase in sophistication it’s exciting to see where ADAS and the future of autonomy are heading.
New technologies are transforming the automotive industry, with innovation redefining what it means to drive or be driven.
The Future of Automotive
As we move deeper into the 21st century, transport and mobility are undergoing two changes that are having a significant impact on the entire world: vehicle electrification and vehicle autonomy.
IMG Series4 NNA
IMG Series4 is a ground-breaking neural network accelerator for the automotive industry to enable ADAS and autonomous driving.