• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2021년도 춘계학술대회
• /
• pp.189-192
• /
• 2021

항적추적 기술에 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용하는 연구로서 기존의 항적추적기술의 경우, 항공기의 등속, 등가속, 급기동, 선회(3D) 비행 등 비행 특성에 따른 칼만 필터 기반의 LMIPDA를 활용한 실시간 항적 추적 시 등속, 등가속, 급기동, 선회(3D) 비행 가중치가 자동으로 변경된다. 이러한 과정에서 등속 비행 중 급기동 비행과 같이 비행 특성이 변경될 때, 항적 손실 및 항적 추적 성능이 하락하여 비행 특성 가중치 변경성능을 향상시킬 필요성이 있다. 본 연구는 레이더의 오차 모델이 적용된 시뮬레이터의 Plot과 표적을 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용하여 학습시키고, 칼만 필터를 활용한 항적추적 결과와 딥러닝 기반 LSTM(Long Short-Term Memory) 모델을 적용한 항적추적결과를 비교함으로써 미리 비행 특성의 변경과정을 예측하여 등속, 등가속, 급기동, 선회(3D) 비행 가중치변경을 신속하게 함으로써 항적추적성능을 향상하기 위한 연구이다.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2006년도 International Symposium on GPS/GNSS Vol.1
• /
• pp.277-282
• /
• 2006

It is well known that Kalman filtering is an optimal real-time data fusion method for GPS/INS integration. However, it has some limitations in terms of stability, adaptability and observability. A Kalman filter can perform optimally only when its dynamic model is correctly defined and the noise statistics for the measurement and process are completely known. It is found that estimated Kalman filter states could be influenced by several factors, including vehicle dynamic variations, filter tuning results, and environment changes, etc., which are difficult to model. Neural networks can map input-output relationships without apriori knowledge about them; hence a proper designed neural network is capable of learning and extracting these complex relationships with enough training. This paper presents a GPS/INS integrated system that combines Kalman filtering and neural network algorithms to improve navigation solutions during GPS outages. An Extended Kalman filter estimates INS measurement errors, plus position, velocity and attitude errors etc. Kalman filter states, and gives precise navigation solutions while GPS signals are available. At the same time, a multi-layer neural network is trained to map the vehicle dynamics with corresponding Kalman filter states, at the same rate of measurement update. After the output of the neural network meets a similarity threshold, it can be used to correct INS measurements when no GPS measurements are available. Selecting suitable inputs and outputs of the neural network is critical for this hybrid method. Detailed analysis unveils that some Kalman filter states are highly correlated with vehicle dynamic variations. The filter states that heavily impact system navigation solutions are selected as the neural network outputs. The principle of this hybrid method and the neural network design are presented. Field test data are processed to evaluate the performance of the proposed method.

• 한국컴퓨터정보학회논문지
• /
• 제26권12호
• /
• pp.11-18
• /
• 2021

본 논문에서는 함포 사격 정확도를 향상시키기 위해 표적 위치 예측과 사격 오차 도출에서의 딥러닝 알고리즘 적용 가능성을 연구하였다. 표적 위치 예측 시 딥러닝 알고리즘의 하나인 LSTM 모델과 RN 구조를 적용했을 때 좀 더 정밀한 표적 위치를 예측할 수 있다는 가능성을 확인하고 모델을 설계하였다. 사격 오차 도출 시 사격제원 계산에 영향을 끼치는 요소들을 데이터 셋으로 관리하며, GAN을 사용하여 데이터 셋을 생성 후 강화 학습을 진행하여 사격 오차를 줄일 수 있는 모델을 설계하였다. 2가지 모델을 결합하여 사격 정확도를 향상시키기 위한 딥러닝 기반의 사격제원 계산 모델을 설계하였다.

• 로봇학회논문지
• /
• 제16권4호
• /
• pp.336-344
• /
• 2021

This paper presents a robust deep learning-based human tracking framework in crowded environments. For practical human tracking applications, a target must be robustly tracked even in undetected or overcrowded situations. The proposed framework consists of two parts: robust deep learning-based human detection and tracking while recognizing the aforementioned situations. In the former part, target candidates are detected using Detectron2, which is one of the powerful deep learning tools, and their weights are computed and assigned. Subsequently, a candidate with the highest weight is extracted and is utilized to track the target human using a Kalman filter. If the bounding boxes of the extracted candidate and another candidate are overlapped, it is regarded as a crowded situation. In this situation, the center information of the extracted candidate is compensated using the state estimated prior to the crowded situation. When candidates are not detected from Detectron2, it means that the target is completely occluded and the next state of the target is estimated using the Kalman prediction step only. In two experiments, people wearing the same color clothes and having a similar height roam around the given place by overlapping one another. The average error of the proposed framework was measured and compared with one of the conventional approaches. In the error result, the proposed framework showed its robustness in the crowded environments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권8호
• /
• pp.2529-2551
• /
• 2022

Visual servoing (VS) based on the Kalman filter (KF) algorithm, as in the case of KF-based image-based visual servoing (IBVS) systems, suffers from three problems in uncalibrated environments: the perturbation noises of the robot system, error of noise statistics, and slow convergence. To solve these three problems, we use an IBVS based on KF, African vultures optimization algorithm enhanced extreme learning machine (AVOA-ELM), and fuzzy logic (FL) in this paper. Firstly, KF online estimation of the Jacobian matrix. We propose an AVOA-ELM error compensation model to compensate for the sub-optimal estimation of the KF to solve the problems of disturbance noises and noise statistics error. Next, an FL controller is designed for gain adaptation. This approach addresses the problem of the slow convergence of the IBVS system with the KF. Then, we propose a visual servoing scheme combining FL and KF-AVOA-ELM (FL-KF-AVOA-ELM). Finally, we verify the algorithm on the 6-DOF robotic manipulator PUMA 560. Compared with the existing methods, our algorithm can solve the three problems mentioned above without camera parameters, robot kinematics model, and target depth information. We also compared the proposed method with other KF-based IBVS methods under different disturbance noise environments. And the proposed method achieves the best results under the three evaluation metrics.

• 전기학회논문지
• /
• 제57권10호
• /
• pp.1861-1868
• /
• 2008

In this paper, Passive Telemerty RF Sensor System using Unscented Kalman Filter algorithm(UKF) is proposed. General Passive Telemerty RF Sensor System means that it should be "wireless", "implantable" and "batterless". Conventional Passive Telemerty RF Sensor System adopts Integrated Circuit type, but there are defects like complexity of structure and limit of large power consumption in some cases. In order to overcome these kinds of faults, Passive Telemetry RF Sensor System based on inductive coupling principle is proposed in this paper. Because passive components R, L, C have stray parameters in the range of high frequency such as about 200[KHz] used in this paper, Passive Telemetry RF Sensor System considering stray parameters has to be derived for accurate model identification. Proposed Passive Telemetry RF Sensor System is simple because it consists of R, L and C and measures the change of environment like pressure and humidity in the type of capacitive value. This system adopted UKF algorithm for estimation of this capacitive parameter included in nonlinear system like Passive Telemetry RF Sensor System. For the purpose of obtaining learning data pairs for UKF Algorithm, Phase Difference Detector and Amplitude Detector are proposed respectively which make it possible to get amplitude and phase between input and output voltage. Finally, it is verified that capacitive parameter of proposed Passive Telemetry RF Sensor System using UKF algorithm can be estimated in noisy environment efficiently.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권12호
• /
• pp.3242-3265
• /
• 2023

Ensuring reliability of a machinery system involve the prediction of remaining useful life (RUL). In most RUL prediction approaches, noise is always considered for removal. Nevertheless, noise could be properly utilized to enhance the prediction capabilities. In this paper, we proposed a novel RUL prediction approach based on noise injection and a Kalman filter ensemble of modified bagging predictors. Firstly, we proposed a new method to insert Gaussian noises into both observation and feature spaces of an original training dataset, named GN-DAFC. Secondly, we developed a modified bagging method based on Kalman filter averaging, named KBAG. Then, we developed a new ensemble method which is a Kalman filter ensemble of KBAGs, named DKBAG. Finally, we proposed a novel RUL prediction approach GN-DAFC-DKBAG in which the optimal noise-injected training dataset was determined by a GN-DAFC-based searching strategy and then inputted to a DKBAG model. Our approach is validated on the NASA C-MAPSS dataset of aero-engines. Experimental results show that our approach achieves significantly better performance than a traditional Kalman filter ensemble of single learning models (KESLM) and the original DKBAG approaches. We also found that the optimal noise-injected data could improve the prediction performance of both KESLM and DKBAG. We further compare our approach with two advanced ensemble approaches, and the results indicate that the former also has better performance than the latters. Thus, our approach of combining optimal noise injection and DKBAG provides an effective solution for RUL estimation of machinery systems.

• 방송공학회논문지
• /
• 제24권3호
• /
• pp.495-505
• /
• 2019

딥 러닝의 대표 알고리즘에는 영상 인식에 주로 사용되는 CNN(Convolutional Neural Networks), 음성인식 및 자연어 처리에 주로 사용되는 RNN(Recurrent Neural Networks) 등이 있다. 이 중 CNN은 데이터로부터 자동으로 특징을 학습하는 알고리즘으로 특징 맵을 생성하는 필터까지 학습할 수 있어 영상 인식 분야에서 우수한 성능을 보이면서 주류를 이루게 되었다. 이후, 객체 탐지 분야에서는 CNN의 성능을 향상하고자 R-CNN 등 다양한 알고리즘이 등장하였으며, 최근에는 검출 속도 향상을 위해 YOLO(You Only Look Once), SSD(Single Shot Multi-box Detector) 등의 알고리즘이 제안되고 있다. 하지만 이러한 딥러닝 기반 탐지 네트워크는 정지 영상에서 탐지의 성공 여부를 결정하기 때문에 동영상에서의 안정적인 객체 추적 및 탐지를 위해서는 별도의 추적 기능이 필요하다. 따라서 본 논문에서는 동영상에서의 객체 추적 및 탐지 성능 향상을 위해 딥 러닝 기반 탐지 네트워크에 칼만 필터를 결합한 방법을 제안한다. 탐지 네트워크는 실시간 처리가 가능한 YOLO v2를 이용하였으며, 실험 결과 제안한 방법은 기존 YOLO v2 네트워크에 비교하여 7.7%의 IoU 성능 향상 결과를 보였고 FHD 영상에서 20 fps의 처리 속도를 보였다.

• 제어로봇시스템학회논문지
• /
• 제2권2호
• /
• pp.108-114
• /
• 1996

We present proofs of the stability and convergence of Self-organizing feature map (SOFM) neural network with time-invarient learning rate and binary reinforcement function. One of the major problems in Self-organizing feature map neural network concerns with learning rate-"Kalman Filter" gain in stochsatic control field which is monotone decreasing function and converges to 0 for satisfying minimum variance property. In this paper, we show that the stability and convergence of Self-organizing feature map neural network with time-invariant learning rate. The analysis of the proposed algorithm shows that the stability and convergence is guranteed with exponentially stable and weak convergence properties as well.s as well.

• 로봇학회논문지
• /
• 제16권1호
• /
• pp.49-55
• /
• 2021

Recently, a smart farm technology is drawing attention as an alternative to the decline of farm labor population problems due to the aging society. Especially, there is an increasing demand for automatic harvesting system that can be commercialized in the market. Pre-harvest crop detection is the most important issue for the harvesting robot system in a real-world environment. In this paper, we proposed a real-time tomato instance tracking algorithm by using deep learning and probability models. In general, It is hard to keep track of the same tomato instance between successive frames, because the tomato growing environment is disturbed by the change of lighting condition and a background clutter without a stochastic approach. Therefore, this work suggests that individual tomato object detection for each frame is conducted by YOLOv3 model, and the continuous instance tracking between frames is performed by Kalman filter and probability model. We have verified the performance of the proposed method, an experiment was shown a good result in real-world test data.