• Smart Media Journal
• /
• v.13 no.5
• /
• pp.9-18
• /
• 2024

Multi-Object trajectory modeling is a major challenge in MOT. CenterTrack tried to solve this problem with a Heatmap-based method that tracks the object center position. However, it showed limited performance when tracking objects with complex movements and nonlinearities. Considering the degradation factor of CenterTrack as the dynamic movement of pedestrians, we integrated the EKF into CenterTrack. To demonstrate the superiority of our proposed method, we applied the existing KF and UKF to CenterTrack and compared and evaluated it on various datasets. The experimental results confirmed that when EKF was integrated into CenterTrack, it achieved 73.7% MOTA, making it the most suitable filter for CenterTrack.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.22 no.6
• /
• pp.251-257
• /
• 2021

An underwater platform operating a passive sonar needs to acquire the target position to perform its mission. In an environment where sea-floor reflections exist, the position of a target can be estimated using the difference in the arrival time between the signals received through multipaths. In this paper, a method of localization for passive sonar is introduced, based on the EKF (Extended Kalman Filter) using the multipath time difference of arrival in underwater environments. TMA (Target Motion Analysis) requires accumulated measurements for long periods and has limitations on own-ship movement, allowing it to be used only in certain situations. The proposed method uses an EKF, which takes measurements of the time differences of the signal arrival in multipath environments. The method allows for target localization without restrictions on own-ship movement or the need for an observation time. To analyze the performance of the proposed method, simulation according to the distance and depth of the target was performed repeatedly, and the localization error according to the distance and water depth were analyzed. In addition, the correlation with the estimated position error was assessed by analyzing the arrival time difference according to the water depth.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.13 no.2
• /
• pp.157-167
• /
• 2010

Unscented Kalman filter based backward filter is derived and the positions from extended Kalman filter, unscented Kalman filter, and extended Kalman smoother are compared and analyzed through a simulation test. Considering the poor GPS signal reception, the simulation is performed under the assumption of only the start and end points of the trajectory, composed of 4 curves and 5 straight sections in the area of $40m{\times}40m $, are known. The test shows that the smoothers generate much better positioning results of 8~9m improvement compared to those from the forward filters. For the comparison between the smoothers, the analysis is performed separately for the curves and straight segments. In both cases, the unscented Kalman smoother generates better positioning error; 10cm and 23cm improved positioning results in straight segment and curves, respectively.

• Journal of IKEEE
• /
• v.17 no.4
• /
• pp.440-445
• /
• 2013

This paper proposes a local estimation system which combines Cell Divide Algorithm with low-cost GPS/INS fused by Extended Kalman Filter(EKF) for localization of Quad-rotor when it returns to the departure point. In the research, the low-cost GPS and INS are fused by EKF to reduce the local error of low-cost GPS and the accumulative error of INS due to continuous integration of sensor error values. When the Quad-rotor returns to the departure point in the fastest path, a moving path can be known because it moves straight, where Cell Divide Algorithm is used to divide moving route into the cells. Then it determines the closest position of data of GPS/INS system fused by EKF to obtain the improved local data. The proposed system was verified through comparing experimental localization results obtained by using GPS, GPS/INS and GPS/INS with Cell Divide Algorithm respectively.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.38 no.4
• /
• pp.345-354
• /
• 2014

In this paper, we propose a real-time algorithm for estimating the relative position and velocity of a person with respect to a robot using a Kinect sensor and an extended Kalman filter (EKF). Additionally, we propose an algorithm for controlling the robot in the proximity of a person in a variety of modes. The algorithm detects the head and shoulder regions of the person using a histogram of oriented gradients (HOG) and a support vector machine (SVM). The EKF algorithm estimates the relative positions and velocities of the person with respect to the robot using data acquired by a Kinect sensor. We tested the various modes of proximity movement for a human in indoor situations. The accuracy of the algorithm was verified using a motion capture system.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1352-1355
• /
• 1997

This paper presents the algorithm for estimating the attitude of an underwater vehicle using EFK. The system model is designed by linerizing the nonlinear Euler angle differential equation and the measurements is a speed logger output. The simulation result shows that the estimation lagorithm is adequate for decreasing attitude errors that grow abruptly during the motion with acceleration and rotation. It also shows that we can adapt the algorithm for compensating initial attitude errors generated after initial leveling.

• Proceedings of the KOR-KST Conference
• /
• 1996.12a
• /
• pp.17-40
• /
• 1996

본 연구에서는 거시적 연속교통류 모형에 바탕을 둔 도시고속도로로 시뮬레이션모형의 개발이 시도되었다. 이 모형은 simple continuum model에 통행수요모형 기능을 강화시킨 것으로서 기존 연속교통류 시뮬레이션모형의 단점을 개선하였다. 제안된 시뮬레이션모형은 정산과정을 거쳐 미국 도시고속도로에서 수집한 현장자료에 의해 평가되었다. 특히 링크의 목적지별 차량대수 추정, 실시간 O-D추정의 문제가 확장칼만필터의 형태로 접근되었으며, 개발된 시뮬레이션모형을 ATMS전략에 활용하는 방안이 개발·평가되었다. 이들 전 과정을 통합한 모수적응적 모형(Parameter Adaptive Model)에 의해 교통량을 실시간으로 예측(Real time traffic prediction)하는 방법을 제안하였으며 현장자료에 의해서 평가되었다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.611-615
• /
• 1991

A technique for estimation of the hydrodynamic parameter of an underwater vehicle is presented. An extended, augmented Kalman Filter is used to extract the hydrodynamic parameter. Computer generated data were used for the measurement information in lieu of actual run data. The feasibility of identifying values of the hydrodynamic parameter of an underwater vehicle is studied. Computer simulation are done in order to validate the performance of the proposed algorithm.

• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10a
• /
• pp.457-462
• /
• 1990

Lately, at an epock of full-scale satellite ranching plan of Korea, T.T.C (Tracking, Telemetery & Command) is a indispensable part. In this paper, particular attention is given to orbit determination problem of role of T.T.C. Orbit determination, which is applied to Kalman Filter and Smoothing Filter, use the observation data which is given by satellite tracking radar system, and then the simulation is accomplished. As a result, it shows effectiveness.

• Journal of Advanced Navigation Technology
• /
• v.11 no.4
• /
• pp.378-387
• /
• 2007

In this paper, Authors develop and verify the algorithm for enhancing the performance of Unmanned vehicle's Autonomous navigation, and also propose the method of establishing much more precise Navigation locus. Unmanned vehicle has a destination, however orientation is not notified, which make it find the future orientation itself. Extended Kalman Filter make it access to the desirable direction, which coupled with INS and GPS is proposed in this paper. Sliding mode control could overcome the side slip and lateral minor movement of the vehicle. The test result would shows the effectiveness of Extended kalman filter and Slide mode control for the navigation.