ADAS: The technology driving automotive safety forward

Advanced driver assistance systems make cars safer and smarter, helping carmakers achieve their vision for a collision-free future

05 Oct 2021

When we empower carmakers to build advanced driver assistance systems (ADAS) that are easy and cost-effective to implement, automotive safety features become more mainstream and more affordable in vehicles.

It only takes a second: A text message on your phone or a child in your back seat distracts you from the road at just the wrong time and, before you know it, your car has traveled several feet into the back of the vehicle that stopped unexpectedly in front of you.

Or it might have, if your automated emergency braking system had not detected the obstacle ahead and applied the brakes within a matter of milliseconds.

Fully self-driving cars may still be a thing of the future, but the safety impact of automation is already being felt. Automotive safety features that would have been considered premium add-ons in the past – such as adaptive cruise control or blind-spot monitoring – are becoming standard as ADAS technology becomes more affordable to implement in vehicles and becomes mainstream in more models.

"People are not necessarily great drivers in mundane situations, but we’re really good at the unpredictable corner cases that computers currently struggle with," said Curt Moore, general manager for Jacinto™ processors. "This makes creating fully automated cars difficult today, but in the meantime, there’s this great opportunity to make cars a lot safer by adding capabilities, like automatically keeping you in your lane or sensing when to stop and go in a traffic jam."

Smarter cars, fewer collisions

ADAS technology enables cars to take actions similar to a driver – sensing weather conditions, detecting objects on the road – and make decisions in real time to improve safety. ADAS features can include automatic emergency braking, driver monitoring, forward collision warning and adaptive cruise control.

Collectively, this technology has the potential to prevent as many as 2.7 million collisions a year in the U.S. alone.1

 

"Going forward, we’re looking at how a car could dynamically change its speed limit not only by location, but also by analyzing the surrounding conditions – such as whether it’s raining or how crowded the road is," said Miro Adzan, sector general manager for ADAS at our company. "We are definitely going to see a significant increase in the level of autonomous decisions a car can make, and that is something that will help reduce collisions and fatalities."

Radar on a chip

More than a decade before modern driver assistance technology began to spread across the auto industry, a small group of engineers at Kilby Labs – our company’s incubator for breakthrough innovation – were working on the technology that would make this possible.

In order to respond to hazards, cars need to both classify obstacles and accurately judge their distance in any weather or lighting conditions. One way to do this is through radar technology. The problem was that radar technology had historically been large, expensive, power-hungry, and designed to detect large objects like planes or tanks – not individual pedestrians.

"We started out on this journey by putting a lot of thought, effort and investment into integrating a complete radar system into a single chip," said Yariv Raveh, TI radar general manager. "This is not something that had previously been done by any other company."

The result was TI mmWave radar sensor, a high-frequency radar sensor the size of a coin. Its resolution can not only detect the presence of an obstacle, but also recognize the difference between a car and a pedestrian and fulfill the latest automotive safety standards, such as the New Car Assessment Program (NCAP), on a single chip.

By combining this with an integrated digital signal processor (DSP), our mmWave radar sensor chip enables machine vision algorithms for object classification to be run directly on the chip itself. This integration simplifies the task of the ADAS manufacturer and helps create a system with low power consumption, small enough to fit in a rear bumper, and at a nominal cost.

Enabling the all-seeing automobile

While the affordability and efficiency of TI mmWave technology is helping radar become a critical part of ADAS, the different strengths of a variety of sensors, such as the high resolution of lidar or color information from cameras, are necessary to enable sufficient all-around reliability. Our FPD-Link™ serializers and deserializers, which transmit uncompressed video data within automotive systems, enable higher-resolution cameras for adaptive driver assistance systems and infotainment displays in the vehicle. This has helped proliferate camera safety features such as front-view camera for collision avoidance, pedestrian recognition, rear-view camera for backup protection and surround-view cameras for parking assistance.

graphic of car at light

By fusing together information from multiple sensing modalities – also known as sensor fusion – the vehicle can help drivers navigate dangerous situations, including in heavy traffic areas and hazardous weather conditions.

“Sensor fusion helps you build a more comprehensive picture of your environment by giving you information from different dimensions,” Miro said. “The sensor data is communicated to a central processing unit, where a machine learning algorithm will use it to interpret the environment. From there, a command can be sent to your braking system, for example, and help prevent a collision.”

Safety-first design

The primary concern in the design of any ADAS system is safety. That’s why processing is critical. As vehicles grow smarter – connecting, communicating, monitoring and making decisions that help prevent accidents – the computing power required to process the enormous amounts of data that make these advanced safety features possible has skyrocketed.

Our Jacinto processors improve awareness of the car’s surroundings and accelerate data-sharing in the software-defined car, efficiently managing multilevel computing in real time while reducing power-hungry memory transfers.

Beyond the processor architecture itself, extensive functional safety documentation and certification of the development process by independent assessment body TÜV SÜD provide manufacturers with information they need to ensure their systems meet the required functional safety standards.

“Automotive safety is important to every driver on the road,” Miro said. “Whether it’s radar, sensor fusion, embedded processing, power management or communication, TI technologies will continue helping carmakers work toward their vision for a collision-free future.”

A passion to create a better world

Helping carmakers make vehicles safer is just one way our innovators are living our company’s passion to create a better world by making electronics more affordable through semiconductors. Each generation of innovation builds upon the last to make technology smaller, more efficient, more reliable and more affordable. We think of this as Engineering Progress. It’s what we do and have been doing for decades.

1.) https://aaafoundation.org/potential-reduction-in-crashes-injuries-and-deaths-from-large-scale-deployment-of-advanced-driver-assistance-systems/ 

Latest company blog posts
04 Apr 2024

Rich and Mary Templeton have given two gifts totaling $91 million to their alma mater, Union College, to transform its engineering, computer science and liberal arts programs and help recruit more women into technology careers

01 Apr 2024

How the design trend of zone architecture improves automotive communication

20 Mar 2024

Why 45- to 130-nanometer process nodes matter

View all blog posts