GPS, INS, Vehicle Kinematics 및 Vision을 활용한 차량용 항법 시스템의 설계

장우진 충북대학교 2017 국내석사

In this paper, integrated GPS(Global Positioning System), INS(Inertial Navigation System), Vehicle Kinematics, and Vision navigation system using loosely coupled method for vehicle navigation is proposed. A distributed EKF(Extended Kalman Filter) is designed to combine the position, the velocity, and the attitude information from a low-cost off-the-shelf GPS receiver, INS, DR(Dead reckoning) using vehicle kinematics model, and vision navigation system. Because of cumulative error, DR and INS are cannot use long time. Because GPS positioning solution is not drift, GPS can use long time. However, because of weak signal power, cannot use in tunnel or urban canyon. To overcome disadvantage GPS, DR and INS single navigation system, integration navigation system is required. DR using vehicle kinematics navigation system is implemented with in-vehicle measurements such as speed from each wheels, and steering angle. Optical flow algorithm is implemented using vision sensor mounted on bumper of vehicle. Also, INS is implemented with 6-axis MEMS(Micro Electro Mechanical System) IMU(Inertial Measurement Unit) sensor. The proposed navigation system is evaluated using RTK(Real Time Kinematics) navigation system.

GPS를 이용한 사진기준점 측량에 관한 硏九

윤부열 東亞大學校 産業大學院 2004 국내석사

최근 동적 DGPS 위치결정에 의한 사진기준점 측량은 국가기본도 작성과 지적재조사 사업을 비롯한 각종 국토개발 분야에 활용할 수 있으며, 앞으로 GPS 수신기 보급의 증가와 더불어 촬영시 비행동체의 정확한 움직임을 고려한 보간기법의 개발, 관성측량기에 의한 GPS의 단점보완 등에 관한 지속적인 연구가 요청되고 있다. GPS를 이용한 항공사진측량 방법에서는 항공기에 탑재한 GPS 수신기로 촬영 순간의 노출점의 위치와 회전각을 정확히 결정함으로써 외부표정요소 결정을 위한 지상기준점 측량을 생략하거나 또는 최소한의 기준점만을 사용하여 외부표정요소를 결정할 수 있으며, 이에 따라 지형도 제작에 대한 시간과 경비를 크게 절감할 수 있다. 본 연구에서는 과천과 수원에 실험지역을 설치하여 이들 실험지역에 대하여 전통적 항공사진측량방법과 GPS 항공사진측량방법을 동시에 수행하고 두 방법의 결과를 비교하여 이들 방법간의 정밀도 및 효율성 등을 평가하고자 하였다. 연구결과 전통적 항공사진측량에 비하여 GPS 항공사진측량의 다양한 장점을 확인할 수 있었다. 특히, 외부표정에 필요한 지상기준점의 수는 약 70%∼80%이상, 지형도제작 비용은 약 30%∼50%이상 절감할 수 있었다. 뿐만 아니라 기준점의 수를 크게 생략하였음에도 불구하고 정밀도에는 전혀 영향이 없었음을 알 수 있었다. Nowadays, GPS-photogrammetry can be applied to the basemap production, a land register and NGIS. And from now on, as the increase of GPS receiver rate, the study on the interpolation methods considering the exact movement of an aircraft at photoflight and the study on the supplement of GPS defect by INS are required continuously. GPS-Photogrammetry, which are based on the direct measurement of the projection centers and attitude at the moment of camera exposure time through loading the GPS receiver in aircraft. This photogrammetric methods can offer us to acquire the exterior orientation parameters with only minimum ground control points, even the ground control process could be completely skipped. Consequently, we can drastically reduce the time and cost for the mapping process. In this thesis, two test flights were conducted in Gwacheon and Suwon area to evaluate the performance of accuracy and efficiency through the analysis of results among the three photogrammetric methods, that is, traditional photogrammetry, GPS-Photogrammetry. Test results shows that a large variety of advantages GPS-Photogrammetry against traditional Photogrammetry is to be verified. Especially, the number of ground control points for the exterior orientation could be saved more than 70∼80%, and the cost for map production 30∼50%, respectively. In addition, it was convinced that the large reduction of control points has not any effect on the block accuracy.

Analysis and Simulation of GPS Software Receiver

장위 전남대학교 대학원 2010 국내석사

GPS(Global Positioning System)is a satellite based navigation system. It is more and more popularly used in every-day life. The modern standard GPS receivers are commonly based on ASIC for signal processing. The main strengths of using ASIC for signal processing include its speed and sensitiveness. However, the downside is the cost of re-designing, and it is unable to re-programmed. The GPS software receiver, which is different from the ordinary receiver, performs all the operations in a programmable microprocessor, making the system highly flexible. So we contra pose the actuality of mobile communication and put forward a new defined wireless communication system called Software based GPS Receiver(SGR). It is the third transform in the wireless communication after simulation to digital and immobility to mobile. Software radio advantages of more effectively saving hard resource, more conveniently and more flexible system, and also in cases where the GPS signals change. In this thesis, the principles and realizations of GPS signals and acquisition of GPS signal. This thesis discussed improved algorithm of Modified acquisition circular correlation. The thesis proposed the simulation results further confirm accuracy and feasibility of the proposed method, this to further studies under the true environment in the simulation platform the GPS signal. Experiments show that the inaccuracy of the carrier frequence estimated within a few hertz by the fine frequence estimation method, and the fine frequence attained can be directly used for the tracking loop. GPS(Global Positioning System)은 위성을 기반으로 한 항법 시스템인데, 이는 일상생활에서 널리 사용되고 있습니다. 현재 사용되고 있는 표준 GPS 수신기는 일반적으로 ASIC에 기초해서 신호처리를 하고 있습니다. 그 이유는 속도가 빠르고 수신감도가 좋기 때문입니다. 그러나 설계를 다시 할 경우에 비용이 많이 들고, 다시 프로그램 할 수 없습니다. 일반 수신기와 다른 GPS 소프트웨어 수신기는 프로그램머블 마이크로 프로세서를 사용하고 있는데, 이는 매우 유용성이 있습니다. 그러므로 앞으로 Software 기반의 GPS 수신기를 사용하는 무선통신에 사용될 것입니다. 이 시스템은 무선이동 통신에 큰 변화를 가져올 것입니다. 소프트웨어를 사용해서 무선통신을 실행함으로서 하드웨어를 절약할 수 있고, 더욱 명확하고 다양한 시스템을 구현할 수 있을 것 입니다. 본 논문에서는 GPS 신호를 실현하고 신호를 수신하는 문제를 해석했습니다. 그리고 개선된 Acquisition circular correlation 알고리듬을 제안합니다. 이 알고리듬을 사용해서 모의실험을 했습니다. 실제 상황과 같은 환경에서 실험을 한 결과, 정확성과 유용성을 확인할 수 있었습니다.

좌표변환시 기준점 수와 분포에 따른 변환특성 분석

박보연 朝鮮大學校 大學院 2004 국내석사

GPS measurement is one of three dimensional location determination methods using satellite and its accuracy and economy have been proved, it has been frequently used for almost measurement works. However, when GPS is applied to current control point measurement, it may result in difficulty that GPS results cannot be used immediately because frame of reference used in GPS is essentially different from that in Korea. Therefore conversion coefficient to convert GPS results into current ones is needed. This study compares and analyzes the characteristics of conversion coefficients according to the number and arrangement of control points used to conversion for calculation of conversion coefficients needed for changing GPS results into current ones and suggests how to structure basic network used for conversion in planning observatory network of GPS measurement. In planning observatory network for GPS measurement, it divides large areas in Gwangju and Jeonnam using trigonometric point, three reference points in Gwangsan-gu and small areas around it using 26 trigonometric points and makes observatory network of various forms. It calculates 7 and 6 variables conversion using the network and compares and analyzes the number of trigonometric points and distribution having the best results. When GPS observatory network is built, it is found that equal distribution of control points to be used for conversion to outer block at a regular distance from reference points is more effective than current methods.

GPS에 의한 신설점의 정밀도 향상을 위한 최적 기준점 선정

김성준 朝鮮大學校 大學院 2002 국내석사

Recently, 3-dimentional topography information is easy to acquire and the accuracy in being improved by using satellite imagery or GPS. Control Point survey by GPS can save labor and time and can obtain result of high accuracy, but if excessive error is contained to Control Point that is used to GPS relative positioning, the error that has influence on new point is propagated to various configuration therefore poor triangular points that are caused by careless management are existed, a product of the triangular points still are in many case. In order to select new point, this study carried out error-analysis of surrounding-control-points and performed selective process of control point. the propriety parts of control points that are used to verify precision is identified by means of comparing existing product with or GPS product about some control points after directly observe each baseline by using EDM. As the error is analyzed by static positioning the optimum control point is selected, the precision of survey data by EDM is improved than GPS data.

GPS를 이용한 항공삼각측량의 정확도 향상에 관한 연구

곽두호 동아대학교 산업대학원 1997 국내석사

This study deals with the airborne kinematic GPS positioning for aerotriangulation. It can decrease the cost which in necessary for the setting up and the preservation of the control points through the use of aerial control points, substitutes for ground control points which is required for the block adjustment with the exact determination of the camera exposure station at the instance of photo flight. This papers introduce the theory of GPS code and the additional parameters required to correct the systematic errors in the combined block adjustment. Also we intend to upgrade the reliability by the robust estimation method which can detect a few gross errors at a time. For example, it is showed that only 1 control point can determine the positions through the test of GPS - aided aerotriangulation at the Wurster watt, the coastland on the north of Germany. But the result was not good because of the low accuracy of C/A-code receiver. However, it is expected that aerotriangulation map production will be effective thanks to the fast improvement of C/A-code receiver and the development of observation method at the coastland which is difficult to set up the control points.

GPS/INS/CDMA망 복합 항법시스템을 위한 통합 필터 설계

김현수 建國大學校 大學院 2003 국내박사

The Need for the positioning indoors as well as outdoors is rising fast in these days due to the U.S. FCC mandate for the E911 and the emerging location based service(LBS) in wireless communication network. There are several viable methods are available. Among them, the most possible and already deployed method, especially in the CDMA communication network, is so called hybrid-positioning method using network assisted GPS and pilot phase TDOA measurements between base and mobile station. For this purpose, a GPS embedded cellular phone using the Qualcomm GPSOne capable modem is commercially available and location service is in service in several countries including Korea, U.S. and Japan. But there are several practical problems under cellular phone user environments inducing large positioning error. Non-Line-of-Sight (NLOS) pilot phase measurement, repeaters in CDMA network, and weak GPS signal inside building are still demanding problem. These induced a large and unpredictable range measurement error. The errors can all be treated as a biased NLOS measurement error. The NLOS mitigation is important for wireless location using CDMA network to acquire the available range information for positioning. A new method for NLOS mitigation using INS and INS-aided GPS is proposed in this thesis. Under benign signal reception environment such as weak signal power, fast vehicle dynamic, interference, and high multi-path, it is hard to track GPS signal. Often it looses GPS signal or produces large pseudorange error while hardly tracking signal. Integrating GPS with INS can resolve this problem. Traditional GPS/INS integration designs use a loosely or a tightly coupled architecture in which the GIPS receiver channels independently track the broadcast satellite signal and constructs the estimated range and range rate to the satellite. In tightly coupled system, the pseudorange rate information is added into the tracking loop of the receiver. This architecture is sub-optimal from the standpoint of preventing signal loss due to jamming and high vehicle dynamics. A deeper level of integration technique is called ultra tightly coupling. The ultra-tightly coupled architecture combines the receiver signal tracking loops and the IMU and the navigation filter function into a single integrated filter. In this case, the filter operates on the receiver-tracking loop I and Q signals to estimate navigation information. In this thesis, we suggested the method for the ultra-tightly coupled GPS/INS integration based on the federated Kalman filter. Finally, we proposed a new method for the integrated navigation using GPS/INS/CDMA network. This method contains of the ultra-tightly coupled GPS/INS integration using federated Kalman filter, the adaptive location filtering algorithm and INS aided NLOS mitigation for CDMA pilot phase measurements, performance evaluation was performed by simulator developed by ourselves. The simulator based on Matlab generates the user's true path and motion data, IF signals of all satellite in view, IMU motion data, the pilot phase measurement for CDMA network. All of simulation parameters were acquired from real experimental data and sensor specifications. The simulator provides the results of tracking loop errors; dynamics stress errors, thermal noise errors, position accuracy and probability density function of pseudorange tracking errors.

GPS 정보를 이용한 위치 및 속도 결정

김현수 建國大學校 大學院 1996 국내석사

GPS(Global Positioning System), the well-known navigation system, is designed by the Department of Defense of America for the first time, which is a non-inertial navigation system that the user who has GPS receiver receives the signal from GPS satellites which rotate 20,200km orbit above and calculates his absolute position in ground. In general, the positioning accuracy of GPS is about loom in 2D rms. Because this accuracy is too big to use for navigation, it needs to consider various navigation algorithms to reduce position error. Also when the number of visible satellites is smaller than four by geographical effects like building blocking in downtown area and mountains blocking, GPS receiver can not calculate navigation solution. It can to a fatal problem for car navigation which will be mainly used in a urban. Therefore, to overcome this shortage of GPS, it can be considered to integrate with other navigation systems. We consider the DR(Dead Reckoning) navigation system. DR system can provide relatively accurate position, but it has a shortage that position error is accumulated continuously with respect to time. Finally, We can implement the cost-effect integrated navigation system which has the algorithm that the shortage of GPS is covered by DR and the error components of DR is compensated by GPS. This thesis, as a basic research to implement the integrated navigation system mentioned above, is focused on developing the algorithms which can provide accurate position, velocity and heading angle using GPS informations to compensate DR errors. For positioning, the least square technique, the Kalman filtering method and a simple differential GPS algorithm are analyzed and tested. For velocity and heading angle determination, algorithm using position rate, Kalman filtering method, the algorithm using Doppler frequency, and the algorithm using relative velocity between satellites and GPS receiver are suggested, derived and tested. And this paper contains a lot of experimental data.

GPS 실내 측위를 위한 도움방식 미약신호 검출 기법 : Assisted weak signal detection method for GPS indoor positioning

부성춘 건국대학교 2004 국내석사

한글초록:GPS는 원래 미국에서 군사 방어 용도로 개발되었다. 그러나 현재 민간인에게도 개방되어 사용되어지고 응용 제품이 개발되어 지고 있다. GPS의 가장큰 응용은 무선 전화의 E911서비스이다. SA가 없을 경우 GPS는 FCC가 요구하는 정확도를 만족시키지만 하드웨어 사이즈나 파워 소모가 사용을 제약하고 있다. 또한 기존 사용 GPS의 경우 미약신호 환경에서 동작을 하지 않는다. 현재 실내나 도심지의 측위 능력이 요구 되어 지고 있다. 이런 응용분야는 신호대 잡음비가 낮은 환경에서 GPS성능에 관심을 갖는다. 또한 낮은 신호에 대해 소프트웨어 상관기를 이용해서 보다 유연성있게 처리하고자 한다. 신호 획득에 필요한 모든 신호 초리는 디지털 신호 처리 기술을 이용하여 소프트웨어 적으로 처리되어 진다. 먼저 이 논문은 Assist GPS에 대해 설명한다. 그 다음으로 빠른 코드 신호 획득, 받은 GPS신호에 대한 DFT 상관기 알고리즘에 대해 설명 한다. 또한 비동기 동기식 방법으로 Assist GPS를 구현 성능 평가를 해보았다.두번째로 미약신호 검출하기 위해서는 긴 적산 시간이 필요하다. 적산시 가장문제가 되고있는 것이 데이터 비트인데 이는 망에서 도움정보를 받아 효과적으로 제거 가능하다. 이런 문제외에 도플러나 클럭에러가 신호 획득에 영향을 줄수 가 있다. 따라서 본 논문에서는 이를 분석하고 효과적으로 보상하였다. 세번째로 실내에서 측위시 항상 문제가 되는 다중경로 오차에 대해 MUSIC이라는 방법을 이용하여 1칩인내의 다중경로 신호를 분리해 낼수 있음을 보여 주었다.마지막으로 구현된 알고리즘의 성능을 평가하기 위해 상용중인 GPS 폰과 실험을 하였다. 실험 결과 구현된 알고리즘이 성능이 GPS 폰보다 좋음을 볼수가 있다 영문초록:GPS was originally developed by the US Department of Defense for military use But it was opened to civilian and applications have been developed. One of the latest applications in GPS is the E911 call service for wireless phones. In this case, current GPS without SA will meets the positioning accuracy requirements by the FCC. But the hardware size and power consumption of GPS receiver limits its usage. In addition to conventional GPS receiver does not operate in the weak signal environment. In particular, indoors and urban positioning capabilities is a requirement. These applications have revived the interest in performance of GPS systems in cases where the signal to noise ratio is very low. Also, software correlator is needed the flexibility. The signal processor takes the sampled IF signal from the RF section of GPS receiver. All Processing required for signal acquisition are implemented in software using the digital signal processing techniques. First, this thesis presents the Assist GPS design. In order to achieve fast code acquisition, correlation of the incoming GPS signal is performed using the DFT method. Because of Doppler shift due to the GPS satellite and user movement, it is impossible to take accurate code delay measurement simply from the code acquisition result. After code acquisition, to reduce the Doppler shift effect and to increase the resolution, effective signal processing techniques are proposed Second, in order to acquire weak signal indoor, we need to increase the integration time. And we should select the best integration method. This thesis presents the integration time increment problem and solution. For example, code Doppler, carrier Doppler, cross correlation interference, clock stability etc. Finally, we discuss the multipath problem. This error is dominant problem in the positioning indoor. And we use the super resolution technique, MUSIC(multiple signal classification algorithm) as multipath mitigation technique. This method is used to improve the resolution beyond that of the traditional PN correlation method. It is able to surpass the traditional Fourier resolution when the channel’s impulse response is modeled as a number of discrete impulses. The higher resolution results in improved ranges enabling location estimates to be made more accurately indoor

도심지에서의 GPS 가시성 분석에 관한 연구

이길재 서울시립대학교 2009 국내석사

Abstract Recently, survey technologies are Computerized, and various detailed and effective survey methods are introduced. GPS-related study also has become one of great interests. The use of GPS is extensively popular due to the improvements in accuracy of information. Getting detailed data is critical in the survey field. However, there is some problem of GPS regarding irregular reception caused by the limit in satellite signal in Urban areas, where it is surrounded by obstructions. In this study, through a simulation and an actual observation, we compared and made an analysis of the condition of reception of GPS survey in the urban areas of Seoul concentrating on Dilution Of Precision(DOP) and visible satellite. This study investigated the actual condition concerning the analysis of GPS’s capability. After selecting an area for experiment to make an analysis, the study has identified the visibility of GPS comparing both the result of Simulation experiment using the KLIS-RN data based on GIS and VRS-GPS equipment. The result of the analysis says that utilizing simulation is certified to be useful during the planning stage of GPS observation since the test results of the simulation the field test are very close. Because of the different result of DOP the limits of property information to be applied for simulations is recognized and the number of visible satellite was decreased by observation location due to obstacle in the skies of urban area. Generally, the number of survey spots tends to be increased in the afternoon rather than in the morning. As an alternative method the GPS survey can be utilized for the basic cadastral survey. Key word : GPS, Visible Satillite, DOP, Simulation 최근 측량방법 및 기술적인 환경이 전산화되면서 정밀하고 효율적인 측량기법의 연구가 다양하나 그중 GPS관련 분야가 활발한 움직임을 보이고 있다. GPS는 위치 정보의 정확도가 향상되면서 다양한 분야에서 활용 범위가 급속하게 확대되고 있으며, 측량 분야는 정밀한 데이터 획득이 중요시 되고 있으나 고층 건물이 밀집한 도심지에서는 위성 신호 수신의 제한으로 GPS 수신 상태의 불규칙으로 인한 문제점이 제기 되고 있다. 본 연구에서는 서울 도심지역의 GPS 측위의 수신 상태를 가시위성과 정밀도 저하율(DOP)에 초점을 맞추어 시뮬레이션과 실제 관측을 통하여 비교, 분석하였다. 본 연구는 이론적으로는 도심에서의 시뮬레이션을 통한 GPS 가용 성능 분석에 관한 실태를 살펴보았고, 실증적 분석을 위해 실험지역을 선정한 후 GIS 기반의 KLIS-RN(도로명 주소 관리시스템) 속성 자료를 통한 시뮬레이션 실험 결과와 VRS-GPS장비로 실제 관측한 결과를 서로 비교, 분석하여 도심지내에서의 GPS 가시성을 확인하는 것으로 하였다. 분석 결과 실험지역에서의 위성 수에 대한 2차례 시뮬레이션 실험과 필드 실험의 결과가 유사한 형태로 도출되어 GPS 관측 계획시 시뮬레이션의 활용이 유용한 것으로 확인하였다. 정밀도 저하율은 상이한 결과가 도출되어 시뮬레이션에 적용되어지는 속성 정보의 한계를 알 수 있었고, 도심의 상공 장애로 인해 관측점별 가시위성의 수는 감소되었다. 전체적으로 실험 대상 지역에 대해 오전 보다는 오후에 측위 가능 지역이 많아지는 것으로 나타났으며, GPS 측위를 달리 하는 방법으로 지적기초측량의 부분에 GPS 측위 활용이 가능할 것이다. 주요어 : GPS, 가시위성, 정밀도 저하율, 시뮬레이션