• 제목/요약/키워드: trajectories

검색결과 1,442건 처리시간 0.027초

Anomalous Trajectory Detection in Surveillance Systems Using Pedestrian and Surrounding Information

  • Doan, Trung Nghia;Kim, Sunwoong;Vo, Le Cuong;Lee, Hyuk-Jae
    • IEIE Transactions on Smart Processing and Computing
    • /
    • 제5권4호
    • /
    • pp.256-266
    • /
    • 2016
  • Concurrently detected and annotated abnormal events can have a significant impact on surveillance systems. By considering the specific domain of pedestrian trajectories, this paper presents two main contributions. First, as introduced in much of the work on trajectory-based anomaly detection in the literature, only information about pedestrian paths, such as direction and speed, is considered. Differing from previous work, this paper proposes a framework that deals with additional types of trajectory-based anomalies. These abnormal events take places when a person enters prohibited areas. Those restricted regions are constructed by an online learning algorithm that uses surrounding information, including detected pedestrians and background scenes. Second, a simple data-boosting technique is introduced to overcome a lack of training data; such a problem particularly challenges all previous work, owing to the significantly low frequency of abnormal events. This technique only requires normal trajectories and fundamental information about scenes to increase the amount of training data for both normal and abnormal trajectories. With the increased amount of training data, the conventional abnormal trajectory classifier is able to achieve better prediction accuracy without falling into the over-fitting problem caused by complex learning models. Finally, the proposed framework (which annotates tracks that enter prohibited areas) and a conventional abnormal trajectory detector (using the data-boosting technique) are integrated to form a united detector. Such a detector deals with different types of anomalous trajectories in a hierarchical order. The experimental results show that all proposed detectors can effectively detect anomalous trajectories in the test phase.

초등학교 수학교과서 속 과제의 학습 경로 분석: 넓이와 부피를 중심으로 (How do Korean Elementary Textbooks Pace Students' Learning to Learning Trajectories? : Focus on Area and Volume)

  • 고은성;이은정;황지현
    • 대한수학교육학회지:수학교육학연구
    • /
    • 제27권2호
    • /
    • pp.207-225
    • /
    • 2017
  • 학생의 학습기회를 형성하는 데 있어 교과서는 상당히 중요한 역할을 한다. 이를 고려하면 교과서 분석은 학생들에게 어떠한 학습 기회를 제공하는지 이해하는데 있어 중요하다. 본 연구는 초등학교 수학교과서를 학습 경로(learning trajectory)에 제시된 넓이와 부피 개념의 이해의 발달 단계에 비추어 분석하였다. 1~6학년의 수학교과서 속의 넓이와 부피에 관한 과제들을 그 과제를 해결하는데 요구되는 사고와 행동을 기초로 학습 경로에 제안 된 발달 단계를 이용하여 코딩하였다. 그 결과, 수학교과서 속의 과제와 학습 경로 간의 상당한 불일치를 발견하였다. 우선, 교과서는 학습 경로에 제안 된 나이보다 상당히 늦게 해당 학습 기회를 제공하였다. 또한, 넓이의 경우 5학년, 부피의 경우 6학년에 학습 기회가 집중되어 있었으며, 이런 집중된 학습 기회도 개념의 발달보다는 넓이 또는 부피의 공식을 적용하는데 중점을 두었다. 본 연구의 결과는 교과서 개발뿐만 아니라 수학수업에서 학생의 학습 기회를 확장시키는데 중요한 시사점을 제시한다.

PSCF 모델을 활용한 부산지역 PM10의 발생원 추정 (Estimation of PM10 source locations in Busan using PSCF model)

  • 도우곤;정우식
    • 한국환경과학회지
    • /
    • 제24권6호
    • /
    • pp.793-806
    • /
    • 2015
  • The purpose of this study is to find out the air flow patterns affecting the PM10 concentration in Busan and the potential sources within each trajectory pattern. The synoptic air flow trajectories are classified into four clusters by HYSPLIT model and the potential sources of PM10 are estimated by PSCF model for each cluster from 2008 to 2012. The potential source locations of PM10 are compared with the distribution of PM10 anthropogenic emissions in east Asia developed in 2006 for the NASA INTEX-B mission. The annual mean concentrations of PM10 in Busan decreased from $51ug/m^3$ in 2008 to $43ug/m^3$ in 2012. The monthly mean concentrations of PM10 were high during a spring season, March to May and low during a summer season, August and September. The cluster2 composed of the air trajectories from the eastern China to Busan through the west sea showed the highest frequency, 44 %. The cluster1 composed of the air trajectories from the inner Mongolia region to Busan through the northeast area of China showed the second high frequency, 26 %. The cluster3 and 4 were composed of the trajectories originated in the southeast sea and the east sea of Busan respectively and showed low frequencies. The concentrations of in each cluster were $47ug/m^3$ in cluster1, $56ug/m^3$ in cluster2, $42ug/m^3$ in cluster3 and $37ug/m^3$ in cluster4. From these results, it was proved that the cluster1 and 2 composed of the trajectories originated in the east and northeast area of China were the causes of high PM10 concentrations in Busan. The results of PSCF and CWT model showed that the potential sources of the high PM10 concentrations were the areas of the around Mongolia and the eastern China having high emissions of PM10 from Beijing, Hebei to Shanghai through Shandong, Jiangsu.

LiDAR 센서를 활용한 배회 동선 검출 알고리즘 개발 (An Algorithm of Identifying Roaming Pedestrians' Trajectories using LiDAR Sensor)

  • 정은비;유소영
    • 한국ITS학회 논문지
    • /
    • 제16권6호
    • /
    • pp.1-15
    • /
    • 2017
  • 최근 국제적인 테러 위협이 불특정 다수를 대상으로 발생하고 있으며, 이러한 위협에서 시민을 보호하기 위한 다양한 대책이 논의 중이다. 저렴해진 센서 기술을 활용한 사전 감시 시스템에 대한 요구가 높아지고 있으나, 보행 궤적의 고유 특성 검출 및 상세 분석 연구가 미비한 실정이다. 본 연구에서는 상용화된 보행 동선 솔루션을 활용하여, 삼성역 개찰구에서 코엑스와 직접 연결되는 연결 통로 (3-6번 출구 근처) 일대의 보행 동선 궤적 조사를 수행하였다. 조사된 궤적 자료를 바탕으로, 궤적 자료의 정규화 기법, Clustering 방법을 중심으로 보행 궤적을 유형화하고 배회 동선을 추출하는 분석 방법론을 제시하였다. 분석 결과, 동일 군집내에서 유사성이 크게 떨어지는 보행 궤적의 검출 가능성을 검증하였다.

대용량 GPS 궤적 데이터를 위한 효율적인 클러스터링 (An Efficient Clustering Algorithm for Massive GPS Trajectory Data)

  • 김태용;박보국;박진관;조환규
    • 정보과학회 논문지
    • /
    • 제43권1호
    • /
    • pp.40-46
    • /
    • 2016
  • 도로지도 생성은 인공위성 촬영이나 현장실사를 기반으로 한다. 그리하여 도로지도를 생성하고 수정하는데 많은 시간과 비용이 든다. 이러한 이유로 차량 GPS 데이터를 이용해 도로지도를 생성하는 연구가 활발히 진행되고 있다. 도로지도 생성 연구에서 가장 중요한 문제는 주도로와 같은 대표궤적을 추출하는 것이다. 대표궤적 추출을 수행할 때에는 시작과 끝이 비슷한 궤적데이터들의 집합을 전제로 하여 궤적을 추출한다. 따라서 대표궤적을 추출하기에 앞서 전처리 과정으로 궤적 클러스터링 작업이 필요하다. 본 논문에서는 이러한 문제를 해결하기 위해 하나의 영역을 일정한 격자로 분할하고, Sweep Line 알고리즘을 응용해 유사궤적들을 탐색한다. 마지막으로 프레쉐거리를 이용하여 궤적 간 유사도를 계산하였다. 실제로 서울의 강남구 지역에 있는 500대의 차량 GPS 궤적을 가지고 클러스터링 작업을 수행하였다. 또한, 실험을 통하여 격자분할 접근방식의 빠른 수행시간과 안정성을 보였다.

불확실한 로보트 시스템의 제어와 파라미터 추정을 위한 반복학습제어기법 (Control and Parameter Estimation of Uncertain Robotic Systems by An Iterative Learning Method)

  • 국태용;이진수
    • 대한전기학회:학술대회논문집
    • /
    • 대한전기학회 1990년도 추계학술대회 논문집 학회본부
    • /
    • pp.421-424
    • /
    • 1990
  • An iterative learning control scheme for exact-tracking control and parameter estimation of uncertain robotic systems is presented. In the learning control structure, tracking and feedforward input converge globally and asymptotically as iteration increases. Since convergence of parameter errors depends only on the persistent exciting condition of system trajectories along the iteration independently of length of trajectories, it may be achieved with only system trajectories of small duration. In addition, these learning control schemes are expected to be effectively applicable to time-varying parametric systems as well as time-invariant systems, for the parameter estimation is performed at each fixed time along the iteration. Finally, no usage of acceleration signal and no in version of estimated inertia matrix in the parameter estimator makes these learning control schemes more feasible.

  • PDF

새로운 슬라이딩 라인을 갖는 가변구조 방식에 의한 직류 모터의 위치 제어 (DC Motor Position Control Using Variable Structure Systems with a New Sliding Surface)

  • 이정훈;이대식;이만고;이주장;윤명중
    • 대한전기학회논문지
    • /
    • 제40권1호
    • /
    • pp.39-46
    • /
    • 1991
  • In the VSS control, the trajectories with a conventional sliding surface have a reaching phase which is an interval from the initial state to the first touching of the sliding surface. Since the sliding mode control can not be realized in a reaching phase, the trajectories may be sensitive to the disturbances and parameter variations. A simple nonlinear sliding surface is proposed to improve the robustness in a reaching phase. The position control of a PM DC servo motor using a new sliding surface is carried out and is compared to the one using the conventional surface. The sliding mode occurs in entire trajectories with the proposed new sliding surface and the improved robustness is obtained.

  • PDF

이족 보행 로봇을 위한 추적 제어 (Tracking Control for Biped Robot)

  • 이용권;박종현
    • 한국정밀공학회:학술대회논문집
    • /
    • 한국정밀공학회 1995년도 추계학술대회 논문집
    • /
    • pp.315-318
    • /
    • 1995
  • In this paper, an optimal trunk trajectory for stable walking of biped robots is expressed as a simple differential equation, which is then solved by numerical methods. We used ZMP (Zero Moment Point), the virtual total ground reaction point within the region of the supporting food, as the criterion of stability of biped robot walking. If the ZMP is located outside of the stable region in dynamic walking, biped robots fall down. The biped robot considered in this paper consists of two legs and a trunk. The trajectories of the two legs and the ZMP of the biped robot are determined such that they are similar ti those of a human. Based upon those trajectories, the trunk trajectory is solved by numerically integrating differential dynamic equations. Leg motions are controlled by the computed torque control method. The effectiveness of control algorithm as well as the trajectories is confirmed by computer simulations.

  • PDF

등엔트로피 궤적에 의한 황사의 장거리 이동 경로 분석 (Isentropic Analysis for the Long Range Trajectories of Yellow Sands)

  • 윤순창;박경선
    • 한국대기환경학회지
    • /
    • 제7권2호
    • /
    • pp.89-95
    • /
    • 1991
  • Yellow sands often occur in Korea during April and May each year, and they are believed to come from the Mongolian Gobi desert as the snow starts to melt in Spring time. Since the analysis of aerosol particulates can hardly distinguish the origin of particulates, the isentropic analysis of meteorological data is often used for the trajectories of the long range transport of yellows sand or air pollutants. The yellow sand case of April 9 $\sim$ 15, 1988, in Korea is analyzed for the identification of long range transport of yellow sands and their trajectories in East Asia, using isentropic analyses. We have tranformed the ECMWF grid data, analyzed in pressure coordinates, into the isentropic coordinates and then have traced the 286 K and 290 K air mass which started Gobi desert. The result shows the transport of yellow sands from the Gobi desert to the Korean peninsula.

  • PDF

Trajectory Planning of Satellite Formation Flying using Nonlinear Programming and Collocation

  • Lim, Hyung-Chu;Bang, Hyo-Choong
    • Journal of Astronomy and Space Sciences
    • /
    • 제25권4호
    • /
    • pp.361-374
    • /
    • 2008
  • Recently, satellite formation flying has been a topic of significant research interest in aerospace society because it provides potential benefits compared to a large spacecraft. Some techniques have been proposed to design optimal formation trajectories minimizing fuel consumption in the process of formation configuration or reconfiguration. In this study, a method is introduced to build fuel-optimal trajectories minimizing a cost function that combines the total fuel consumption of all satellites and assignment of fuel consumption rate for each satellite. This approach is based on collocation and nonlinear programming to solve constraints for collision avoidance and the final configuration. New constraints of nonlinear equality or inequality are derived for final configuration, and nonlinear inequality constraints are established for collision avoidance. The final configuration constraints are that three or more satellites should form a projected circular orbit and make an equilateral polygon in the horizontal plane. Example scenarios, including these constraints and the cost function, are simulated by the method to generate optimal trajectories for the formation configuration and reconfiguration of multiple satellites.