Electronics and Semiconductors

Self-Driving Evolves Through Advanced Driver Assistance

26 November 2015

The extensive publicity about the road testing of automated vehicles makes it easy to overlook numerous innovations that are also transforming driving. Collectively known as advanced driver assistance systems (ADAS), these developments by auto manufacturers make cars safer through an evolution in vehicle sensing, intelligence and control that will ultimately lead to affordable self-driving cars.

Advanced electronics take up much of the space in test cars and their expense rivals that of the vehicles themselves. To make self-driving feasible for automobiles in series production, the technology must be small, lightweight and affordable. Moreover, there are legal and social hurdles to be dealt with and overcome. Since these changes can only be accomplished gradually, the development, introduction and conversion of cars on the road to self-driving will be an evolutionary process that spans a number of automotive generations.

Advanced driver assistance provides the environment that fosters these adaptations. By its nature, ADAS perfects different aspects of automated control—at first in independent subsystems, then with increasing levels of system integration, until ultimately the vehicle can drive itself. Enabling technology for this evolution comes from advanced integrated circuit (IC) solutions that provide sensing, communications and intelligence for ADAS systems.

What ADAS Includes

ADAS technologies exist at different levels of active assistance, as shown in the non-exhaustive list of ADAS features below. Some of these are available today, while others are in development for upcoming models. Information and warning systems are passive, leaving the operator in full control at all times. Function-specific and combined-function automation permit the vehicle to operate itself under certain conditions, though the driver can take back control at any time. Features with limited self-driving automation take full control of the vehicle for specific tasks. Each higher level of autonomy employs technologies from the more basic levels, leading up to fully-autonomous cars that will operate on their own, with or without a driver.

Information and warning

• Side and rear cameras with driver displays

• Backup assistance

• Warning of oncoming cross-traffic

• Ultrasound sensing of areas in front and behind the car obstructed from driver view

• Traffic sign recognition

• Detection of lane markings

• Blind spot warning

• Night vision

• Ranging of objects ahead in conditions of poor visibility

• Vehicle-to-vehicle and vehicle-to-infrastructure communication

• Dynamic display image of the entire car and surrounding space

• In-cabin monitoring and warning to a distracted driver

Function-specific automation

• Lane keeping assist to stay in the center of the lane

• Active cruise control for changing traffic conditions

• Collision avoidance by automatic braking

• Automated emergency brake

• Intelligent head beam assistance that illuminates curves ahead or prevents blinding of other traffic participants

Combined-function automation

• Adaptive cruise control with lane centering

• Traffic jam assist

Limited self-driving automation

• Automated parking/valet parking

• Highway auto pilot

Full self-driving automation

Typically, advanced features appear initially in high-end models, and then migrate to mid-range vehicles and eventually to all new cars. Where safety is concerned, insurance companies, regulatory bodies and legislatures often accelerate phase-ins through favorable premiums and legal mandates. Sometimes legislation has to resolve questions of liability or change existing requirements. For example, external rear-view mirror mandates have now been changed in many countries in favor of streamlined, no-blind-spot cameras and interior displays. As cars begin to communicate with each other and with installations along the roadway, security from malicious intrusion will also be a concern. Issues such as these involve legal and social adaptations, as well as functional safety standards that cannot happen overnight.

Technical Requirements for ADAS

ADAS technology evolves by shrinking individual components and integrating functions into more comprehensive systems. Highly automated ADAS features will fuse data inputs from different sensor elements, including multiple cameras, ultrasound, LIDAR and radar. More sensors and higher-resolution cameras call for high-bandwidth, low-latency communications and high-performance processing. Optimized networking that allows video and other data, control signals and power over a single wire will significantly reduce weight, size and installation time.

Visual processing forms the bulk of high-level ADAS work because of the massive amount of pixel data. The system has to condition video images, identify areas of interest, recognize items and then decide what to do. Each of these time-critical tasks serves to narrow the data stream while increasing algorithmic complexity. The system requires heterogeneous processing, ranging from dedicated video signal processing hardware for the raw data input, through programmable digital signal processors (DSPs) for object scanning and recognition, to a high-performance microprocessor for decision-making.

Automotive voltages vary over a wider range than do those of most electronic systems, and the operating environment can be extremely stressful in terms of temperature range, vibration and contamination. Robust power management solutions not only supply reliable power, but also offer important protections that help prevent damage to systems. Given the evolving needs of ADAS, developers require ICs that offer cost-efficiency, flexibility, reliability in the extreme conditions of automotive operation and performance overhead for growth.

An ADAS Vision Processor

At the heart of ADAS systems that are based on camera fusion, as well as some radar sensing, is a high-performance application processor that integrates a range of heterogeneous functions to perform vision-processing tasks effectively. For instance, TI’s TDAxx system-on-chip solutions integrate a general-purpose dual-Cortex-A15 RISC processor and dual C66x DSPs that handle mid- and higher-level algorithms, and a programmable Vision AccelerationPac for specialized image computation. Software support allows fast creation of algorithms and data flows for video capture, preprocessing, analytics and display. Integrated peripherals, including support for communication through a single co-axial cable, simplify design, minimize space and facilitate sensor fusion.

Advanced driver assistance is with us now and is rapidly increasing in importance. Innovative IC solutions, along with legal and social changes, will enable more complex ADAS features and lead to the ultimate success of fully-automated vehicles.

Powered by CR4, the Engineering Community

Discussion – 0 comments

By posting a comment you confirm that you have read and accept our Posting Rules and Terms of Use.
Engineering Newsletter Signup
Get the GlobalSpec
Stay up to date on:
Features the top stories, latest news, charts, insights and more on the end-to-end electronics value chain.
Weekly Newsletter
Get news, research, and analysis
on the Electronics industry in your
inbox every week - for FREE
Sign up for our FREE eNewsletter