메뉴 건너뛰기
소속 기관 / 학교 인증
인증하면 논문, 학술자료 등을  무료로 열람할 수 있어요.
한국대학교, 누리자동차, 시립도서관 등 나의 기관을 확인해보세요
(국내 대학 90% 이상 구독 중)
고객센터 ENG
주제분류

논문 기본 정보

저자정보
(국민대학교) (국민대학교)
저널정보
한국자동차공학회 한국자동차공학회 추계학술대회 및 전시회 2017년 한국자동차공학회 추계학술대회 및 전시회
오류 신고하기

피인용 0

검색

    초록·키워드

    Control performance is significantly affected by two control parameters, sampling frequency and sensing to actuating delay. Higher sampling frequency and shorter delay makes better control performance, however, it makes the system difficult to schedule in a multitasking environment, where multiple control tasks share CPU resource. To maintain high frequency and short delay, task periods and deadlines should be kept very short that eventually leads to system unschedulability. Making the problem more complex, in AUTOSAR-based vehicle control systems, a large number of small functions runnables are mapped to periodic tasks that matches each runnable’s own period and data flow between sensing and actuation goes through a sequence of runnables with different periods. With this motivation, we present a novel algorithm that determines runnable periods that provides both the optimal control performance and system schdulability for AUTOSAR. Assuming that control performance is given as a linear function of frequency and delay, we use Lagrange multiplier to find the set of optimal periods with a constraint that the system utilization is less than 100% with EDF scheduling algorithm.

    최근 본 자료 전체보기