퍼지 이론을 이용한 수중 운동체의 자세계산 혼합 알고리즘

Mixing algorithm for attitude computation of underwater vehicle using fuzzy theory

  • 김영한 (서울대 대학원 제어계측공학과) ;
  • 이장규 (서울대 공대 제어계측공학과) ;
  • 한형석 (순천향대 공대 제어계측공학과)
  • 발행 : 1996.02.01

초록

In this paper, attitude computation algorithm for a strap down ARS(Attitude Reference System)of an underwater vehicle has been studied. Attitude errors o the ARS using low-level gyroscopes tend to increase with time due to gyroscope errors. To cope with this problem, a mixing algorithm of accelerometer aided attitude computation has been developed. The algorithm can successfully bound the error increase for cruising motion, but it gives instantaneously large errors when a vehicle maneuvers. To improve the performance in case of vehicle's maneuver, a new attitude computation mixing algorithm complying state of vehicle and to manage the adjustment of the gains which are invariant in the existing algorithm. In addition, a gain scheduling method is applied to fuzzy inference composition process for real-time computation. Monte Carlo simulation results show that the proposed algorithm provides better performance than the existing algorithm.

키워드