• 한국항공운항학회지
• /
• 제21권1호
• /
• pp.23-29
• /
• 2013

This paper presents a tracking filter of radar beacon system (RBS) for automatic takeoff and landing of an unmanned aerial vehicle. The proposed tracking filter is designed as the decoupled tracking filter to reduce the computational burden. Also, an adaptive estimation method of the measurement error covariance is proposed to provide an improved tracking performance compared to the conventional decoupled tracking filter whenever the accuracy of RBS observations is degraded. 100 times Monte Carlo runs performed to analyze the performance of the proposed tracking filter in case of normal operation and degraded operations, respectively. The simulation results show that the proposed tracking filter provides the improved tracking accuracy in comparison with the conventional decoupled tracking filter.

• 한국ITS학회 논문지
• /
• 제6권3호
• /
• pp.45-56
• /
• 2007

비디오를 이용한 차량 추적 자동화 기술은 차량속도 등의 정확한 교통량 측정을 위해 중요하다. 그러나 단일 카메라로 취득된 비디오를 이용한 차량추적의 경우 카메라의 유한한 시계(field of view)때문에 추적범위가 제약된다. 차량추적 범위를 넓히고 차량행태에 대한 보다 많은 정보를 얻기 위해서는 복수개의 카메라로 연속된 추적영역을 촬영하여 분석하는 방식을 이용할 수 있다. 본 논문에서는 규칙기반 차량추적 알고리즘을 적용한 두 대의 동기된 비디오 시??스에서 연속적으로 추적하는 방식을 제안한다. 제안된 방법에서는 영상분석을 통해 나타날 수 있는 거의 모든 경우를 규칙을 만들기 위해 고려되었으며, 칼만 필터를 이용하여 예측성능을 향상시켰다. 제안된 연속된 비디오에서의 차량추적 방법은 확장된 추적범위에서 차량의 순간속도, 차량의 개별행태 등에 보다 많은 정보를 취득하는데 성공적으로 활용될 수 있다.

• 한국정보전자통신기술학회논문지
• /
• 제10권2호
• /
• pp.133-140
• /
• 2017

본 논문은 컴퓨터 비전 기술을 기반으로 자동 차량 감시 시스템을 제안하였다. 실시간 주행표시 시스템은 ITS의 필수 요건을 충족하면서, 자동 감시제어가 가능한 시스템이다. 이러한 장점은 확실한 자동차 추적에 대해 주요 장애물 처리 시스템 적용할 경우, 움직이는 물체에 대한 그림자 추적이다. 추적 차량 이미지에서 모든 종류의 정보를 획득하기 위해 차량을 확실하게 감시 화면에 나타나게 하였다. 시간이 지남에 따라 차량을 정밀 추적 제어 할 필요가 있고, 입체 모델 기반접근 방식 또한 필요한 방식으로 적용하였다. 일반적으로 개체 또는 와이어 프레임 모델의 골격에 의해 차량의 각각의 유형을 나타내었고, 시스템이 실시간 실행되지 않더라도 차량 궤적은 3D기반 방식에서 높은 정밀도로 측정 될 수 있다는 점을 보여 준다. 본 논문에서는 차량, 배경, 그림자에 적용 가능하고, 도로 교통 감시의 시스템을 분할 방법을 역시 적용하였다. 과속 자동차의 속도 추적을 통해 낮은 레벨의 차량제어 추적기의 유효성 역시 실행 하였다. 결론에서 개발된 추적 제어 시스템에서 향상된 자동차 추적의 방법을 개선하고자 하였으며 고속도로 감시제어 시스템을 개발하고자 하였다.

• Journal of Mechanical Science and Technology
• /
• 제16권2호
• /
• pp.137-146
• /
• 2002

In this paper, a path generation algorithm that uses sensor scannings is described. A scanning algorithm for recognizing the ambient environment of the Automatic Guided Vehicle (AGV) that uses the information from the sensor platform is proposed. An algorithm for computing the real path and obstacle length is developed by using a scanning method that controls rotating of the sensors on the platform. The AGV can recognize the given path by adopting this algorithm. As the AGV with two-wheel drive constitute a nonholonomic system, a linearized kinematic model is applied to the AGV motor control. An optimal controller is designed for tracking the reference path which is generated by recognizing the path pattern. Based on experimental results, the proposed algorithm that uses scanning with a sensor platform employing only a small number of sensors and a low cost controller for the AGV is shown to be adequate for path generation.

• 대한임베디드공학회논문지
• /
• 제13권6호
• /
• pp.321-328
• /
• 2018

This paper presents an automatic target tracking flight system using a PID controller based on velocity command of a multirotor UAV. The automatic flight system includes marker based onboard target detection and an automatic velocity command generation replacing manual controller. A quad-rotor UAV is equipped with a camera and an image processing computer to detect the marker in real time and to estimate the relative distance from the target. The marker tracking system consists of PID controller and generates velocity command based on the relative distance. The generated velocity command is used as the input of the UAV's original flight controller. The operation of the proposed system was verified through actual flight tests using a marker on top of a moving vehicle and tracks it to successfully demonstrate its capability using a quad-rotor UAV.

• 한국ITS학회 논문지
• /
• 제4권2호
• /
• pp.1-11
• /
• 2005

차량의 자동 추적은 도로 건설 계획을 수립하는데 있어서 중요한 자료인 교통의 흐름을 분석하거나, 비디오 영상의 교통량 분석 시스템에서 차량의 정확한 속도를 측정하는 등 여러 분야에서 그 중요성이 날로 증가하고 있다. 본 논문에서는 비디오 영상에서 차량 추적 시 발생할 수 있는 가능한 경우들을 고려한 규칙기반 차량 추적 방법들을 제안한다. 제안된 방법들은 정화도의 관점에서 Mean-Shift방법에 비해 빠르고 좋은 성능을 나타내었다 본 논문에서 제안된 방법은 추적의 속도나 정확성을 높이기 위해 칼만 필터 예측기 및 칼라 정보를 이용한 탐색들을 융합한다면 더 우수한 추적 성능을 보일 것으로 기대된다.

• 드라이브 ㆍ 컨트롤
• /
• 제13권1호
• /
• pp.18-26
• /
• 2016

This research is conducted to develop an automatic steering system for parallel parking, and the performance of the system was evaluated by parallel parking a conventional vehicle. The automatic steering system consisted of MDPS (motor driven power steering) to control steering, ESC (electronic stability control) to acquire wheel speed, ultrasonic sensors to recognize the parking space, and a controller to communicate and handle data. The parallel parking process using the automatic steering control consisted of parking space recognition, parking path generation, and parking path tracking. The path for parallel parking was generated based on a kinematic model of a conventional vehicle, and a PI controller was used to control the steering angle for path tracking. Parallel parking using the automatic steering control was conducted according to vehicle speed conditions. The results show that the errors on the x-axis and y-axis were below 0.54 m and 0.14 m, respectively, and the error on the steering angle was less than $1^{\circ}$. Therefore, it is possible to implement parallel parking using an automatic steering control system for conventional vehicles.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2005년도 ICCAS
• /
• pp.1014-1019
• /
• 2005

In this paper, a neural network technique for automatic steering control of a four wheel drive autonomous highway snowplow vehicle is presented. Controllers are designed by the LQR method based on the vehicle model. Then, neural network is used as an auxiliary controller to minimize lateral tracking error under the presence of load. Simulation studies of LQR control and neural network control are conducted for the vehicle model under a virtual snowplowing situation. Tracking performances are also compared for two and four wheeled steering vehicles.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권2호
• /
• pp.710-726
• /
• 2018

This paper presents an extended particle filter to increase the accuracy and decrease the computation load of vehicle tracking. Particle filter has been the subject of extensive interest in video-based tracking which is capable of solving nonlinear and non-Gaussian problems. However, there still exist problems such as preventing unnecessary particle consumption, reducing the computational burden, and increasing the accuracy. We aim to increase the accuracy without an increase in computation load. In proposed method, we calculate the direction angle of the target vehicle. The angular difference between the direction of the target vehicle and each particle of the particle filter is observed. Particles are filtered and weighted, based on their angular difference. Particles with angular difference greater than a threshold is eliminated and the remaining are stored with greater weights in order to increase their probability for state estimation. Threshold value is very critical for performance. Thus, instead of having a constant threshold value, proposed algorithm updates it online. The first advantage of our algorithm is that it prevents the system from failures caused by insufficient amount of particles. Second advantage is to reduce the risk of using unnecessary number of particles in tracking which causes computation load. Proposed algorithm is compared against camshift, direction-based particle filter and condensation algorithms. Results show that the proposed algorithm outperforms the other methods in terms of accuracy, tracking duration and particle consumption.

• 한국마린엔지니어링학회:학술대회논문집
• /
• 한국마린엔지니어링학회 2000년도 춘계학술대회 논문집(Proceeding of the KOSME 2000 Spring Annual Meeting)
• /
• pp.11-15
• /
• 2000

This paper provides a basic study on automatic of a trailer type vehicle which consists of two parts such as a tractor and a trailer Backward moving and parking control is very important to automate this type of vehicle. However it is very difficult to control such their motion since a trailer type vehicle is a non-holonomic system. Therefore in this paper we propose the backward path tracking control algorithm for a trailer type vehicle. And also this paper presents the results of simulation to verify the effectiveness of the proposed control algorithm.