Driving automation technologies will not only reshape the automotive industry and its related value chain, but also provide a safer and more comfortable travel experience. A number of countries including China, the United States, Japan, and South Korea are currently using driving automation technologies as a starting point to develop intelligent transportation systems on a national scale. In China, the Ministry of Industry and Information Technology (MIIT) has published a national standard for levels of driving automation which will come into effect on January 1, 2021.
The enhanced compute power of AI chipsets and increasingly mature perception-based AI algorithms are enabling active safety and ADAS. Driving automation technologies are expected to explode in popularity over the next 5 to 10 years, heralding the dawn of an entirely new era for the automotive industry.

Advanced Driver Assistance System (ADAS) uses sensors inside and outside the vehicle to obtain real-time information about the environment, identify and track static and moving objects, and perform systematic computing and analysis. With these capabilities, ADAS makes driving safer and more efficient.

The HiSilicon ADAS solution provides a range of high-performance heterogeneous computing platforms, including the NPU, digital signal processor (DSP), intelligent video engine (IVE), data processing unit (DPU), and CPU. With the all-weather 24-hour image signal processor (ISP) and the 4K camera system (single-, dual-, and triple-camera systems are supported), our ADAS can obtain information relating to its surroundings, as well as detect and analyze various traffic participants in real time. Our ADAS then uses computing, prediction, and sensor fusion to deliver a wide range of functions, including:

• Forward collision warnings (FCW)
• Advanced emergency braking (AEB)
• Lane departure warnings (LDW)
• Intelligent speed limit information (ISLI)
• Blind spot detection (BSD)
• Rear collision warnings (RCW)

We will continue to invest in our ISO 26262-compliant chipsets and solutions and improve upon their intelligent compute power and image processing capabilities, ensuring optimal performance as efficient chipset platforms for ADAS.

Highlights of HiSilicon ADAS

• Industry-leading visual perception: Supports 4K and higher resolutions and all-weather 24-hour ISP capabilities
• Advanced AI computing: Supports 4 TOPS of compute power, efficient AI development platforms, and mainstream algorithm frameworks

Automatic Parking Assist (APA) utilizes ultrasonic radars, visual sensors, and ADAS processors to locate appropriate parking spaces. It identifies spaces that are the right size for the vehicle, plans the parking route, and controls the steering wheel, gearbox, and accelerator pedal to assist with parking. APA is more intelligent than traditional parking assist systems, and helps drivers park their vehicles quickly and safely even in narrow spaces.

Automated Valet Parking (AVP) enables drivers to park or summon their vehicles with just one tap. Once the vehicle has received an instruction, it either parks itself or drives from its current parking space to the pick-up location. Drivers are not required to monitor their vehicles throughout these processes, as AVP is a more automated system than APA.

HiSilicon APA and AVP solutions support 8-channel ISP capabilities, and use a short-distance camera and an ultrasonic sensor to display a 360 degree view of the area surrounding the vehicle in real time. With 4 TOPS of AI compute power, the APA and AVP can automatically search for parking spaces and identify hidden obstacles, resulting in safer and more accurate parking.

• Multi-channel video processing: A single chipset supports 4 to 8 camera input channels, and the ISP is independently configured, supporting flexible camera configuration with a resolution of 2 to 8 megapixels and covering all parking scenarios.
• Robust AI compute power and development platform: Our solution provides 4 TOPS of AI compute power, efficient algorithm tuning tools, and device-side self-learning platforms.
• Surround view: Our chipsets are embedded with hardware acceleration modules for image stitching, delivering a three-dimensional 360 degree view of the area surrounding the vehicle in real time.