The automotive industry is evolving from the SAE L2 (the vehicle controls acceleration, braking and steering under human supervision) to the fully autonomous L5 (the vehicle does not need to interact with people), so the demand for powerful image sensors is growing to support advanced driver assistance systems ( ADAS) and various camera systems for autonomous driving.
This article refers to the address: http://
There are many different requirements for the camera system around the car body. Automakers often require front-facing cameras with the highest resolution to drive on highways that need to look far; the next-generation emergency braking system has medium resolution and wide field of view; for side cameras, low- and medium-resolution Yes, to provide 360-degree machine vision for assisted parking, lane change, or advanced autonomy in urban driving.
Evolving ADAS solutions require more functional sensors to meet ever-increasing automotive safety standards. As the degree of auto networking and autonomy increase, the safety and integrity of automotive systems is critical to automakers. The image sensor is the eye of the vehicle, and any interruption or tampering of the sensor function can cause the vehicle to become "scorpion" or mislead decision algorithms. The number of cameras around the body is increasing, each camera is connected to a central system, and data and commands are transmitted through the lines between the sensors and the processor. These systems are more versatile than sensors and processors in a single enclosed space. fragile.
ON Semiconductor's new line of scalable image sensors, the AR0820AT, AR0220AT and AR0138AT, meet the specific resolution requirements of different camera systems around the body. These sensors offer resolutions from 8.3 MP (megapixels) to 1.2 MP with industry-leading micro-lighting performance, functional safety features up to ASIL-C, high dynamic range (HDR) and industry-first network security options.
ADAS and autopilot solution developers now have a range of image sensors with different resolutions but a combination of common architecture and features, enabling developers to reduce development time and development costs. Early development with a single sensor, adapting its algorithm to pixel performance and system characteristics, and then extending it to higher resolution through further testing. This enables the development of high-performance automotive camera systems with the flexibility to deploy cameras of different resolutions in different locations on a single vehicle or to provide different levels of functionality for different vehicles.