부산대학교 도서관

The future of driving extends from physical mobility and enjoyment today to having more services enabled by wireless connectivity, clean and green technologies, secure transactions, and so forth. Whether the ownership is based on physical vehicles or services, or whether the drivers are human or robots, one demand that will never change is to have better safety at affordable cost. Among available sensor technologies for the advanced-driver-assistance-system (ADAS), radar is indispensable due to its unique capability in robustness against environmental impacts, long-range detection, sufficient range resolution, and simultaneous multi-depth detection. Those are very crucial since camera, Lidar and ultrasonic sensors perform poorly under severe weather conditions, and an autonomous vehicle would become partially blinded without radars. There are several automotive radar applications such as front radars responsible for autonomous cruise control and automatic emergency braking, as well as corner radars responsible for blind-spot detection (BSD), cross-traffic alert (CTA), and the like. A new class of applications comprehending ultra-short range sensing and 360° surround view for parking assistance, door-opening alert, etc. is emerging. In this paper, requirements of the new applications will be examined, which will be further broken into system and circuit specification. A new system including application-driven algorithm, hardware and software designs will be presented to fulfill the new demands.