PRISM :: Browsing by Author "Gao, Yang"

Browsing by Author "Gao, Yang"

Now showing 1 - 20 of 70

Open Access
A Java-based wireless framework for location-based services applications
(2002) Liu, Zhe; Gao, Yang
Open Access
A New Cooperative PPP-RTK System with Enhanced Reliability in Challenging Environments
(2023-07) Lyu, Zhitao; Gao, Yang; Wang, Ruisheng; O'Keefe, Kyle; Gao, Yang
Compared to the traditional PPP-RTK methods, cooperative PPP-RTK methods provide expandable service coverage and eliminate the need for a conventional expensive data processing center and the establishment and maintenance of a permanently deployed network of dense GNSS reference stations. However, current cooperative PPP-RTK methods suffer from some major limitations. First, they require a long initialization period before the augmentation service can be made available from the reference stations, which decreases their usability in practical applications. Second, the inter-reference station baseline ambiguity resolution (AR) and regional atmospheric model, as presented in current state-of-art PPP-RTK and network RTK (NRTK) methods, are not utilized to improve the accuracy and service coverage of the network augmentation. Third, the positioning performance of current PPP-RTK methods would be significantly degraded in challenging environments due to multipath effects, non-line-of-sight (NLOS) errors, poor satellite visibility and geometry caused by severe signal blockages. Finally, current position domain or ambiguity domain partial ambiguity resolution (PAR) methods suffer from high false alarm and miss detection, particularly in challenging environments with poor satellite geometry and observations contaminated by NLOS effect, gross errors, biases, and high observation noise. This thesis proposed a new cooperative PPP-RTK positioning system, which offers significant improvements to provide fast-initialization, scalable coverage, and decentralized real-time kinematic precise positioning with enhanced reliability in challenging environments. The system is composed of three major components. The first component is a new cooperative PPP-RTK framework in which a scalable chain of cooperative static or moving reference stations, generates single reference station-derived or reference station network-derived state-space-representation (SSR) corrections for fast ambiguity resolution at surrounding user stations with no need for a conventional expensive data processing center. The second component is a new multi-feature support vector machine (SVM) signal classifier based weight scheme for GNSS measurements to improve the kinematic GNSS positioning accuracy in urban environments. The weight scheme is based on the identification of important features in GNSS data in urban environments and intelligent classification of line-of-sight (LOS) and NLOS signals. The third component is a new PAR method based on machine learning, which employs the combination of two support vector machine (SVM) to effectively identify and exclude bias sources from PAR without relying on satellite geometry. The prototype of the new PPP-RTK system is developed and substantially tested using publically available real-time SSR products from International GNSS Service (IGS) Real-Time Service (RTS).
Open Access
A robust quality control system for GPS navigation and kinematic positioning
(1993) Gao, Yang; Krakiwsky, Edward J.
The thesis describes the development and testing of a robust quality control system for GPS navigation and kinematic positioning. The system is built upon the successful combination of well-known classical statistics with modem robust statistics. Good performance in failure detection and identification, minimum implementation effort as well as automation are the main criteria employed in the development of the system. Theoretical and numerical aspects behind the system are presented along with the introduction of important concepts and the derivation of useful equations. The system was tested by applying it to integrity monitoring in GPS navigation and cycle slip detection and identification in GPS kinematic positioning and results show that significant improvements have been achieved as compared to the conventional approach.
Open Access
A software engine for the rapid development of mobile asset management systems
(2004) Lee, Suen Man; Gao, Yang
Open Access
Ambiguity Resolution with Precise Point Positioning
(2014-06-26) Wang, Min; Gao, Yang
Ambiguity resolution with PPP aims at effectively reducing the long convergence time. The research in this dissertation explores its potential for both dual-frequency signals and triple-frequency signals. One of the most popular PPP ambiguity resolution strategies with dual-frequency GNSS signals is to first fix the L1/L2 wide-lane ambiguities in the geometry-free approach and then fix the L1 ambiguities in the geometry-based approach. A software package has been developed to evaluate the PPP ambiguity resolution performance with this strategy, including position accuracy, time-to-first-fix and fixing availability of both L1/L2 wide-lane ambiguity and L1 ambiguity, etc. A new model has been developed to improve the performance of PPP ambiguity resolution with dual-frequency signals, in which the L1 fractional bias is split into one direction-independent and three directional-dependent components for each satellite. Better performance can be obtained at both server and client rover side using the new model, but the L1 ambiguity fixing time still requires around 30 minutes on average. A new method of instantaneous PPP ambiguity resolution with triple-frequency signals has been proposed, which involves first fixing the L2/L5 wide-lane ambiguities in geometry-free approach and then fixing the L1/L2 wide-lane ambiguities in geometry-based approach. Based on the test results with extensive MATLAB simulation datasets and newly available BeiDou real signal datasets, both L2/L5 wide-lane ambiguity and L1/L2 wide-lane ambiguity can be fixed instantaneously and reliably using a single epoch of triple-frequency measurements. A carrier smooth carrier technique has been proposed to reduce the measurement noise for PPP ambiguity resolution with triple-frequency signals. PPP can achieve horizontal positioning accuracy better than 5 cm and 3D positioning accuracy better than 10 cm, with the convergence time less than two minutes. This performance is comparable to RTK.
Open Access
An Improved Particle Filter Algorithm for Geomagnetic Indoor Positioning
(2018-03-19) Huang, He; Li, Wei; Luo, De An; Qiu, Dong Wei; Gao, Yang
Geomagnetic indoor positioning is an attractive indoor positioning technology due to its infrastructure-free feature. In the matching algorithm for geomagnetic indoor localization, the particle filter has been the most widely used. The algorithm however often suffers filtering divergence when there is continuous variation of the indoor magnetic distribution. The resampling step in the process of implementation would make the situation even worse, which directly lead to the loss of indoor positioning solution. Aiming at this problem, we have proposed an improved particle filter algorithm based on initial positioning error constraint, inspired by the Hausdorff distance measurement point set matching theory. Since the operating range of the particle filter cannot exceed the magnitude of the initial positioning error, it avoids the adverse effect of sampling particles with the same magnetic intensity but away from the target during the iteration process on the positioning system. The effectiveness and reliability of the improved algorithm are verified by experiments.
Open Access
Assessment and Improvement of Underground Coal Gasification Modelling
(2018-12-13) Jiang, Liangliang; Chen, Zhangxin; Farouq-Ali, S. M.; Abedi, Jalal; Jeje, Ayodeji A.; Gao, Yang; Gupta, Subodh C.
Underground coal gasification (UCG) is a process to convert coal in-situ into combustible synthetic gas (syngas). Oxidant is brought downhole through an injector for coal combustion and gasification while resulting syngas is extracted from a producer. UCG offers a better way to exploit coal resources over conventional mining with smaller environmental footprint. It has gained considerable attention in emerging economies, e.g., China and India, which are coal-rich nations and have an ever-increasing energy demand. As a complex coal thermal recovery process involving multi-physics and kinetics, knowledge gaps remain before UCG reaches large-scale commercial implementation. To enhance knowledge, this work applies a modified simulation tool to model certain important aspects of UCG, i.e., assessing critically the use of a reservoir simulator to model UCG, exploring the theory of a prior linking method of reverse combustion, studying the role of coal cleats in governing fluid flow and heat transport with regard to aquifer contamination, and investigating the progressive changes in coal pores associated with UCG. Additional modelling efforts were made to explore an extended practical importance of UCG. The prospect of applying UCG to mobilize contiguous heavy oil is studied and the feasibility of linking UCG with carbon storage and sequestration is examined.
Open Access
Attitude Estimation Methods Using Low-cost GNSS and MEMS MARG Sensors and Their Integration
(2022-09) Ding, Wei; Gao, Yang; El-Sheimy, Naser; Noureldin, Aboelmagd
For low-cost magnetic, angular rate, and gravity (MARG) sensors based on the microelectromechanical system (MEMS) technology, the sensor errors and measurement noises are significantly large. Attitude errors by integrating gyro data accumulate rapidly. When the vehicle is quasi-static, the roll and pitch angles can be determined by accelerometer measurements which use the local gravity as the reference. The magnetometer is resorted to generate heading information by measuring the geomagnetic field. However, the accelerometer and magnetometer measurements can be deteriorated by the vehicle maneuver and ambient artificial magnetic disturbances, respectively. Thereby a quaternion-based error state Kalman filter (ESKF) is developed to fuse the MEMS MARG sensor measurements for accuracy improved attitude estimation. The error state vector constitutes attitude error and gyro bias variation. the gyro-measured angular rates are used to continuously propagate the vehicle’s three-dimensional attitude quaternion in its sampling rate, whilst accelerometer and magnetometer measurements are employed for the state correction. Disturbances such as external accelerations and magnetic anomalies are excluded, and the measurement noise covariance matrix is adaptively adjusted according to the innovations. Global navigation satellite system (GNSS) based attitude estimation shows time-independent error characteristics. The pitch and heading angles can be determined using a single GNSS antenna based on the time differenced carrier phase (TDCP) observations or derived from a moving baseline formed between two firmly mounted GNSS antennas. The major challenges of the former include cycle slips, carrier phase discontinuity, and slow vehicular velocity which should be excluded from attitude estimation. Whereas the integer ambiguity resolution is indispensable for the latter, the baseline length constrained least-squares ambiguity decorrelation adjustment (C-LAMBDA) method can be applied. The GNSS/MARG sensors integrated attitude estimation methods are investigated to exploit the complementary merits of the high precision of MARG sensor during the short period and the performance stability of GNSS over the long term. The ESKF developed for the MARG sensor is extended to utilize the GNSS-derived heading and pitch angles for additional measurement updates. The solution continuity is guaranteed by the MARG sensor alone during the periods when the GNSS-derived attitude angles are unavailable.
Open Access
Carrier phase based ionosphere recovery over a regional area GPS network
(2001) Liao, Xiangqian; Gao, Yang
Open Access
Carrier Phase-Based Ionospheric Modeling and Augmentation in Uncombined Precise Point Positioning (UPPP)
(2018-09-21) Xiang, Yan; Gao, Yang; O'Keefe, Kyle P. G.; Nielsen, Jorgen; Rangelova, Elena V.; Chen, Wu
Precise Point Positioning (PPP) is a stand-alone high-precision positioning technique employing carrier phase measurements and external augmentation or aiding products. PPP reduces labor and equipment costs in contrast to Real-Time Kinematic (RTK) which relies on base stations. However, PPP suffers from a long convergence time of 15 to 60 minutes to reach the centimetre level. This long initialization time restricts the applications of PPP. To address this problem, we make use of accurate and precise ionospheric corrections. This dissertation endeavors to improve the ionospheric observables, Differential Code Biases (DCBs), and Mapping Function (MF). We then leverage these to reduce the convergence time. To obtain more accurate ionospheric corrections, we retrieve ionospheric observables using PPP. The ionospheric observables from the more commonly-used carrier phase smoothed code method are adversely affected by levelling errors. PPP offers a preferable way to reduce the leveling errors and preserve the consistency of ionospheric corrections, beneficial for shortening the convergence time of PPP. We demonstrate that the ionospheric observables retrieved from three PPP models, Traditional Ionosphere-Free, University of Calgary (UofC), and Uncombined (UPPP), all agree in terms of DCBs. The differences of ionospheric observables are at centimetre level. With the improved ionospheric observables using PPP, the stability and internal accuracy of satellite and receiver DCBs are also enhanced. The Root Mean Square (RMS) of the satellite DCB estimates is improved from 0.1 nanoseconds to 0.07 nanoseconds, and the day-to-day stability is enhanced by 0.22 nanoseconds. Another factor affecting ionospheric corrections is the MF which is mostly based on the fixed height Single-Layer Model (SLM). To reduce the effects of the inhomogeneity of the ionosphere, an Ionospheric Varying Height (IVH) is proposed and examined. Results show the mapping errors are reduced by about 15% when the integral varying height is exploited. By applying the improved ionospheric corrections into UPPP, we achieve an accuracy of 0.4 metres for global constraints and 0.2 metres for the regional constraints at the first epoch. The convergence time for the simulated kinematic mode is reduced from 41 to 7.5 minutes in the east at one decimetre, from 14.5 to 4.0 minutes in the north at one decimetre, and from 11.0 to 6.5 minutes in the vertical at two decimetres at a 68% confidence level.
Open Access
Coastal sea level change from satellite altimetry and tide guages
(2012) Tang, Feng; Sideris, Michael G.; Gao, Yang
Abstract In this thesis, in order to obtain accurate sea level anomalies time series from satellite altimetry data in the coastal areas, ocean tide corrections from a local tide model and residual ocean tides derived by the least-squares spectral and harmonic analysis were removed from I 0-day TOPEX/POSEIDON and Jason- I data, which have been corrected for other bias and errors. The variations of altimetry time series reduce from 9.6 cm to 5.2 cm on average in the west coast of Canada and the values reduce from 8.7 cm to 5.1 cm on average in the east coast of Canada. Cleaned altimetry and tide gauge data were submitted to a first-order linear regression model to determine the absolute and relative linear sea level trend for every time series, respectively. The noise model for every time series was tested by the power spectral density and maximum likelihood. The white and, the white and flicker noise model are found to be the best noise model for altimetry and tide gauge data respectively. Our rate estimate results agree with published values very well. GPS-derived rates of vertical land motion were added to the relative sea level trend derived from tide gauge data. A joint analysis method is developed to combine satellite altimetry and tide gauge rate estimates together for estimating the regional sea level change. Regional trend estimates by the joint method are very similar to the results by satellite data only. It indicates that the contribution of tide gauge data in the joint analysis method is very limited.
Open Access
Design, Implementation and Key Issues of Adaptive Tightly Coupled MEMS INS/GPS Integration System
(2017) Zhou, Qifan; El-Sheimy, Naser; Hai, Zhang; El-Sheimy, Naser; Ling, Pei; Hai, Zhang; Noureldin, Aboelmagd; Gao, Yang; Rui, Zhou
Tightly-coupled integrated system is advantageous in providing seamless navigation solution compared with other integration schemes, and is attractive in multiple application fields. This thesis focuses on the design, implementation of adaptive tightly-coupled integrated navigation system with the aiding of external measurement. The key issues investigated and problems solved in this dissertation are: 1. It presents four novel adaptive noise estimation approaches. These methods rely on the observation information to estimate the measurement noise property and error characteristic. The noise assessment method makes use of difference operation to eliminate the effect of other elements and this process is decoupled from the filter calculation loop. Thus, the covariance matrix estimation result will not be coupled with the state vector error, and this existed problem in traditional adaptive Kalman filter is avoided. 2. It proposes a new low-cost MEMS sensor in-filed calibration algorithm. The navigation mission requires calibration before started. The proposed approach can perform calibration of the sensors without any requirement or special needs. The calibration scheme is convenient to be accomplished by simple hand rotation in space. The bias, scale factor error and non-orthogonal error are able to be identified. 3. It investigates the Euler angle based attitude estimation. The Euler angle attitude update is constrained for further practical application owning to its singularity problem. The singularity problem will cause the attitude procedure discontinue and involves more estimation error. An intelligent coordinate switch algorithm is proposed to overcome this drawback and has achieved a good performance. The adaptive noise estimation approach is also applied in the filter to adjust the weight of observation model, which prevents the negative effect of external acceleration. 4. It introduces the adaptive noise estimation theorem and external observation in the standard tightly-coupled integrated system, and establishes the system hardware platform to test the validation. The height and heading information measured by barometer and magnetometer are used involved in measurement model to provide aiding information. A switch filter strategy is designed to save computational time, and the adaptive noise estimation approach is used to acquire the GPS measurement error characteristic. Both simulation and practical tests are conducted to verify the system.
Open Access
Development and Assessment of a Seismic Waveform Capturing System using Precise Point Positioning with High-rate GNSS Observation
(2021-01-20) Jiang, Yang; Gao, Yang; Sideris, Michael G.; Gao, Yang; Sideris, Michael G.; Cheng, Y. Frank; Rangelova, Elena V.
It is of great significance to build a system for earthquake early warning (EEW) and rapid hazard assessment based on real-time seismic waveform capturing. A top-rated geodetic monitoring tool that is the Global Navigation Satellite Systems (GNSS) has been widely used to perform such a task. First, the relative GNSS positioning methodology of the real-time kinematics (RTK) provides high-accuracy trajectory estimations while it has weaknesses such as the influence of base station vibrations. Second, the absolute positioning technique of precise point positioning (PPP) is more suitable for earthquake scenarios, while it relies on accurate satellite products such as orbit and clock corrections. Although real-time free-access correction service is available, it remains unclear whether PPP's performance meets an earthquake capturing system's requirements. On the other hand, most current GNSS-based earthquake capturing systems provide temporal resolution of no greater than 10Hz, limiting the earthquake recording performance, especially for high-frequency seismic wave components up to 100Hz. Moreover, GNSS monitoring stations usually require a high budget for purchase and maintenance. However, the recently modernized GNSS constellations and the creation of cost-effective, high-performance GNSS receivers that are still evolving have created the potential for an affordable and accurate GNSS solution that can be used for seismic waveform capturing. In this thesis, a real-time earthquake capturing system is developed based on PPP with ambiguity resolution (PPPAR) using real-time correction service. As is designed, the system can support the processing of high-rate multi-frequency multi-constellation GNSS observations with sampling rate up to 100Hz. The high sampling rate allows for the extraction of high-frequency seismic wave components, which is important for earthquake detection and rapid assessment. The performance of the developed system has been assessed based on a simulation earthquake experiment with a cost-effective 100Hz GNSS receiver. The performance assessment includes the trajectory estimation accuracy, waveform property accuracy, and the sampling rate's impact. The analysis showed that the system software and algorithm are efficient and fully operational, while centimeter-level accuracy for the trajectory and waveform property is continuously achieved. Furthermore, a real-time earthquake capturing system with a higher sampling rate of GNSS observations can provide more accurate results.
Open Access
Development of a mobile equipment management system
(2000) Ramsaran, Ronald M.; Gao, Yang
Open Access
Development of a New Real-Time Precise Point Positioning System
(2018-03-14) Yang, Hongzhou; Gao, Yang; El-Sheimy, Naser; Kattan, Lina; Skone, Susan H.; Bisnath, Sunil B.
Real-time Precise Point Positioning (PPP) is drawing increasing attentions from both the Global Navigation Satellite System (GNSS) community and real-time users with different applications, such as offshore navigation, precise agriculture and hazard warning. To meet the rapidly increasing demand, the International GNSS Service (IGS) Real-Time Service (RTS) is currently disseminating several real-time high-frequency State Space Representation (SSR) products through the Internet under the Networked Transport of Radio Technical Commission for Maritime Services (RTCM) via the Internet Protocol (NTRIP) protocol. High availability of real-time PPP services requires high availability of precise orbit and clock corrections. Any correction outage, either due to corrupted ephemeris or loss of communication link, will degrade the availability of precise positioning using the service. Meanwhile, the communication burden is very heavy with such high update rates. To tackle the above limitations, a new robust real-time PPP system with higher availability is proposed in this thesis. The proposed system consists of three components regarding server end, communication end and user end. For the new real-time PPP system, the satellite orbit and clock Initial Parameters (IP) products are generated at the server end and broadcast to the user end for the generation of high precision orbit and clock products, afterwards, the real-time PPP can be carried out with the IP-based high precision satellite products. With the IP products, the real-time PPP system can operate with scalable update rates according to the various accuracy requests of different applications. Furthermore, the new real-time PPP system can continue during Internet connection outages, which is not uncommon in real applications due to the Internet connection losses. The prototype of the new real-time PPP system is developed and substantially tested in real-time.
Open Access
Development of laser fluorosensor data processing and gis tools for oil spill response
(2009) Jha, Maya Nand; Gao, Yang
Open Access
Enhancing Land Vehicle Navigation in Challenging Environments Using Consumer Level Devices
(2020-11-20) Moussa, Mohamed; El-Sheimy, Naser; Noureldin, Aboelmagd MA; Moussa, Adel; Helaoui, Mohamed; Gao, Yang; El-Mowafy, Ahmed
Recently there has been a massive effort in developing navigation systems for the self-driving cars. GNSS/INS integration is the most common sensor fusion technique to estimate the land vehicles navigation states. However, this system is not perfect in all operating situations as GNSS signals may suffer from signal outages, and/or multipath in urban and foliage areas. In such cases, INS provides the navigation solution which is degraded after a very short period due to the large drifts of the INS. During GNSS signal outage, INS should be assisted with other aiding sensors to mitigate its large drift. These sensors may include magnetometers, odometers, cameras, Light Detection And Ranging (LIDAR), Radio Detection And Ranging (RADAR), etc. Maps aiding navigation is used in many previous researches to help low-cost INS in GNSS denied environments. Consumer Portable Devices (CPDs) are widely used all over the globe. CPDs contain many sensors that could particpate in the enhancement of the land vehicles. Unfortunately, there are some limitations associated with the previous mentioned aiding techniques related to their high price, high computation and processing cost, weather and surrounding environmental effects in addition to the map aiding method drawback that is based on the avialability and the update rate of the required maps. Therefore, autonomous land vehicles navigation using low-cost sensors integrated systems has attained a lot of research interest The main objective of this research is to develop various land vehicle navigation systems that work in GNSS challenging environment, based on low-cost, non conventional sensors, and CPDs to add a redundant land vehicle motion information, reduce the cost of the navigation systems and thus decrease the overall self-driving car cost and provide an accepatble navigation performance. Two low-cost non-conventional wheel odometry systems are proposed where the first system is based on multiple ultrasonic sensors while the other is developed using multiple low-cost gyroscopes. These systems are implemnted to aid the low-cost INS in GNSS signal outage to reduce its drift. The relative alignment between the CPD and vehicle frames is very vital process when using CPD in the land vehicle navigation. However, it requires other sensors to estimate the relation between the CPD and vehicle coordinaite systems and it should be in motion. A new static relative alignment method is developed based on the vehicle vibration signature pattern which does not require any additional sensors. Moreover, a steering wheel angle estimation method is developed using CPD self-contained accelerometers that is attached to the steering wheel where this information is used in aiding the low-cost INS in GNSS challenging environment. Furthermore, different DR and aiding systems are investegated based on the land vehicle information and the CPD sensors to assess the navigation performance for such systems. Moreover, multiple CPD navigation systems are investegated using federated fusion technique.Two non-conventional navigation systems are proposed where the first depends on the ECU sensors to provide a redundant land vehicle speed information to aid the low-cost INS in GNSS signal outage. The second system is based on mass flow sensors that benefits from the aerodynamics of the vehicle motion to provide the speed and the heading information in indoor environment.The performance of the proposed low-cost navigation techniques show promising navigational results and low complexity efforts.
Open Access
Error analysis and stochastic modeling of MEMS based inertial sensors for land vehicle navigation applications
(2004) Park, Minha; Gao, Yang
Open Access
Evaluation of Kinematic GNSS PPP for Tropospheric Zenith Wet Delay Estimation in Mountainous Regions
(2021-08-31) Gratton, Paul Thomas; O'Keefe, Kyle; Lachapelle, Gérard; O'Keefe, Kyle; Lachapelle, Gérard; Banville, Simon; Gao, Yang
In this research, the effectiveness of kinematic zenith wet delay (ZWD) estimation using global navigation satellite systems (GNSS) precise point positioning (PPP) techniques is evaluated. The major challenges of kinematic ZWD estimation compared to static mode are (1) significant and variable GNSS signal obstruction, (2) trajectory durations of several hours compared to several days in static mode and (3) strong correlation between ZWD and height estimates. High-end and low-cost receivers are tested on vehicular highway trajectories through mountainous regions with height changes over 1000 m and varying levels of GNSS obstruction. Results are compared to static tests with open-sky conditions. Static agreement of ZWD profiles of high-end receivers was found to be at the sub-millimetre level. Agreement of low-cost receivers when using a high-grade antenna was found to be at the level of 3 mm or better. Low-cost receivers using low-cost antennas suffered ZWD biases of 3 cm due to height biases of 7-10 cm. Kinematic accuracy of ZWD profiles for high-end receivers in trajectories with minimal obstruction was found to be 5 mm, increasing to 10 mm in trajectories with more obstruction and 25 mm in very harsh obstructions. Accuracy of ZWD profiles for low-end receivers ranged from 8-20 mm in open conditions and 20-35 mm in more challenging conditions. Low-cost receivers were not tested in very harsh obstructions. Empirical ZWD models were found to agree with high-end receiver PPP-derived ZWD profiles within 15 mm or better, hence accuracy poorer than 15 mm appears ineffective.
Open Access
Floor Plan Based Indoor Vision Navigation Using Smart Device
(2013-07-10) Huang, Bei; Gao, Yang
The Global Positioning System (GPS) nowadays is sized down to a chip sensor and built into almost every smart phone and tablet. Therefore, navigation using those intelligent gadgets becomes a must-have function. GPS has been widely employed for outdoor navigation, while its performance suffers from severe degradation in challenging scenarios such as urban canyon and indoor. Due to the overwhelming signal noise, building reflection and blockage, indoor navigation using GPS frequently encounters poor accuracy or even signal outage. In order to improve the service availability and navigation accuracy, inertial measurement units (IMU) are integrated with GPS, which continuously measures the user acceleration and rotation rate. Integrating these relative motion measurements derives the position, velocity and orientation, therefore it bridges the gap during GPS outage. However, IMU raw measurements are contaminated by sensor bias and drift, and for low-cost sensors on smart devices, the bias and drift are extremely severe and unstable. The navigation solution derived from these poor quality sensors results in significant accumulative errors, which will destroy the system reliability very soon. Furthermore, most smart devices embrace cellular and Wi-Fi network positioning to improve service availability, time-to-first-fix, accuracy and reliability in indoor scenarios. Unfortunately, network based positioning performance highly depends on the signal reception, and quality of the database of Wi-Fi access points (APs) and cellular towers. Based on our experiments, network based positioning performance can merely achieve tens of meters position accuracy on average. For indoor navigation application, however, users’ expectation is room-level, turn-by-turn navigation guidance. In this thesis, a vision navigation system is developed for pedestrian indoor navigation using smart device. In order to derive the three-dimensional camera position from the monocular camera vision, a geo-reference database is needed. Floor plan is a ubiquitous geo-reference database that every building refers to it during construction and facility maintenance. Comparing with other popular geo-reference database such as geo-tagged photos, the generation, update and maintenance of floor plan database does not require costly and time consuming survey tasks. In the proposed system, user is asked to take a picture of the surrounding indoor scenario, and a robust feature matching method is designed to match the indoor features contained in the camera image to those in the floor plan database. Given the image-to-floor plan feature correspondences, a navigation algorithm is developed to integrate the monocular vision with the floor plan geo-reference information and derive the camera position and orientation. The vision navigation system is realized on an iOS App and tested with iPad in various indoor scenarios. The test results show that, comparing with Wi-Fi positioning, the proposed system has improved the position accuracy from tens of meters to 5 m on average.