http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
비선형 상태공간 모델을 위한 Point-Mass Filter 연구
최영권(Yeongkwon Choe) 강원대학교 산업기술연구소 2023 産業技術硏究 Vol.43 No.1
In this review, we introduce the non-parametric Bayesian filtering algorithm known as the point-mass filter (PMF) and discuss recent studies related to it. PMF realizes Bayesian filtering by placing a deterministic grid on the state space and calculating the probability density at each grid point. PMF is known for its robustness and high accuracy compared to other nonparametric Bayesian filtering algorithms due to its uniform sampling. However, a drawback of PMF is its inherently high computational complexity in the prediction phase. In this review, we aim to understand the principles of the PMF algorithm and the reasons for the high computational complexity, and summarize recent research efforts to overcome this challenge. We hope that this review contributes to encouraging the consideration of PMF applications for various systems.
비력벡터매칭 기법을 이용한 자세결정 알고리즘의 성능 향상
최영권(Yeongkwon Choe),박찬국(Chan Gook Park) 한국항공우주학회 2017 韓國航空宇宙學會誌 Vol.45 No.2
항공기 및 지상 이동체 등에 사용되는 자세 및 방위 결정 시스템은 자세를 결정하기 위해 중력가속도 벡터와 지구자기장 벡터를 이용한다. 이를 위해 가속도계와 자력계를 이용하게 되는데, 가속도계의 경우 중력가속도뿐만 아니라 항체의 운동 가속도까지 포함하게 되어 가속 중에는 자세결정이 어려워진다. 본 논문에서 다루는 가속도 보상 방법은 가속도계에서 얻은 비력으로부터 GPS 수신기를 통해 계산한 항체의 가속도를 빼주어 이를 해결하는 방법이다. 기존의 알고리즘은 보상한 벡터를 상수 형태로 간주해 이용하게 되는데, 본 논문에서는 이로 인한 오차를 분석하고 측정치로부터 모델을 재유도해 성능을 개선했다. 기존의 알고리즘이 내포한 오차 요인과 본 논문에서 제안한 알고리즘에 의해 자세 추정 성능이 개선됨을 컴퓨터 시뮬레이션을 통해 확인했다. Attitude determination algorithms for aircraft and land vehicles use earth gravitational vector and geomagnetic vector; hence, magnetometers and accelerometers are employed. In dynamic situation, the output from accelerometers includes not only gravitational vector but also motional acceleration, thus it is hard to determine accurate attitude. The acceleration compensation method treated in this paper solves the problem to compensate the specific force vector for motional acceleration calculated by a GPS receiver. This paper analyzed the error from the corrected vector regarded as a constant by conventional acceleration compensation method, and improve the error by rederivation from measurements. The analyzed error factors and improvements by the proposed algorithm are verified by computer simulations.
프로베니우스 놈 기반 이중 인자 적응 필터를 이용한 INS/GPS 초강결합 기법
최영권(Yeongkwon Choe),강창호(Chang Ho Kang),김선영(Sun Young Kim),박찬국(Chan Gook Park) 제어로봇시스템학회 2018 제어·로봇·시스템학회 논문지 Vol.24 No.8
This paper presents a deeply coupled INS/GPS system that uses a Frobenius norm-based adaptive filter for countering signal interferences. A deeply coupled INS/GPS system integrates an INS (inertial navigation system) and a receiver based on vector tracking loops. Vector tracking loops track signals from satellites based on navigation solutions, which provides great immunity to signal interferences and the high dynamics of a receiver. However, dependence on the navigation solution means that if the navigation solution is wrongly estimated once, the entire tracking loop can fail. Therefore, navigation filters for measurement noise must be properly handled to avoid tracking failures when signal interferences exist. This paper applies a Frobenius norm-based adaptive filter. Whereas conventional Frobenius norm-based adaptive filters use only one adaptive factor, this paper expands the filters to use two adaptive factors to deal with INS/GPS deep integration. To verify the performance of expanded filters, an experiment is conducted with a commercial GPS noise jammer. The test results exhibit better performance compared to conventional methods.