A new image sensor for the automotive and autonomous vehicle market

点击翻译按钮获取中文摘要

Modern automobiles possess an incredible amount of technology, offering advanced monitoring and activesystems to improve safety and convenience, including lane departure warning, blind spot monitoring, dynamiccruise control and automated parking. The next major advance in automobiles is expected to be self-driving(i.e., car drives itself with human supervision) and autonomous (i.e., cars drives itself with no humanintervention) vehicles. While a limited number of self-driving cars are on the market now (e.g., certain Tesla &BMW models), autonomous consumer cars are not yet. However, forecasts suggest annual sales > 500kcars/year in the United States alone by 2025. Autonomous vehicles (AVs) require a lot of sophisticated sensorsto manage the many functions involved in the complicated task of driving. Common is an image sensor(similar to those used in standard cameras) to provide information for object identification, colour, edgedetection, etc.; however these provide very little 3D data. 3D data is usually obtained using LIDAR systems,which are expensive and produce large data streams requiring costly on-board processing capabilities. UBC hasdeveloped a new complementary metal-oxide-semiconductor (CMOS) image sensor (CIS) that providesimportant benefits over current image sensors and provides 3D information. Our smart CIS enables anintegrated, on-chip computational imaging sensor solution that could compete in the automotive market,potentially bridging both the image and LIDAR sensor markets. The objective of this proposal is to conduct aprofessional and detailed market assessment of the potential of using UBC technology in the automotive andautonomous vehicle market.