This thesis describes the theoretical and practical stages through development to testing of an integrated navigation system, specifically composed of an Inertial Navigation System (INS), and Global Positioning System (GPS). Integrated navigation systems combine the best features of independent systems to bring out increased performance, improved reliability and system integrity. In an integrated INS/GPS system, INS output is used to calculate current navigation states / GPS output is used to supply external measurements, and a Kalman filter is used to provide the most probable corrections to the state estimate using both data. Among various INS/GPS integration strategies, our aim is to construct a tightly coupled integrated INS/GPS system. For this purpose, mathematical models of INS and GPS systems are derived and they are linearized to form system dynamics and system measurement models respectively. A Kalman filter is designed and implemented depending upon these models. Besides these, based on the given aided navigation system representation a quantitative measure for observability is defined using Gramians. Finally, the performance of the developed system is evaluated with real data recorded by the sensors. A comparison with a reference system and also with a loosely coupled system is done to show the superiority of the tightly coupled structure. Scenarios simulating various GPS data outages proved that the tightly coupled system outperformed the loosely coupled system from the aspects of accuracy, reliability and level of observability.

Present land vehicle positioning and navigation relies mostly on the Global Positioning System (GPS). However, in urban canyons, tunnels, and other GPS-denied environments, the GPS positioning solution may be interrupted or suffer from deterioration in accuracy due to satellite signal blockage, poor satellite geometry or multipath effects. In order to achieve continuous positioning services, GPS is augmented with complementary systems capable of providing additional sources of positioning information, like inertial navigation systems (INS). Kalman filtering (KF) is traditionally used to provide integration of both INS and GPS utilizing linearized dynamic system and measurement models. Targeting low cost solution for land vehicles, Micro-Electro-Mechanical Systems (MEMS) based inertial sensors are used. Due to the inherent errors of MEMS inertial sensors and their stochastic nature, which is difficult to model, KF has limited capabilities in providing accurate positioning in challenging GPS environments. This research aims at developing reliable integrated navigation system capable of demonstrating accurate positioning during long periods of challenging GPS environments. Towards achieving this goal, Mixture Particle filtering (MPF) is suggested in this research as a nonlinear filtering technique for INS/GPS integration to accommodate arbitrary inertial sensor characteristics, motion dynamics and noise distributions. Since PF can accommodate nonlinear models, this research develops total-state nonlinear system and measurement models without any linearization, thus enabling reliable integrated navigation and mitigating one of the major drawbacks of KF. Exploiting the capabilities of PF, Parallel Cascade Identification (PCI), which is a nonlinear system identification technique, is used to obtain efficient stochastic models for inertial sensors instead of the currently utilized linear models, which are not adequate for MEMS-based sensors. Moreover, this research proposes a method to update the stochastic bias drift of inertial sensors from GPS data when the GPS signal is adequately received. Furthermore, a technique for automatic detection of GPS degraded performance is developed and led to improving the performance in urban canyons. The performance is examined using several road test experiments conducted in downtown cores to verify the adequacy and the benefits of the methods suggested. The results obtained demonstrate the superior performance of the proposed methods over conventional techniques. / Thesis (Ph.D, Electrical & Computer Engineering) -- Queen's University, 2010-07-23 20:27:02.12

Autonomous land vehicles (ALVs) have received considerable attention after their introduction into military and commercial applications. ALVs still stand as a challenging research topic. One of the main problems arising in ALV operations is the navigation accuracy while the other is the dynamic effects of road irregularities which may prevent the vehicle and its cargo to function properly. In this thesis, we propose intelligent solutions to these two basic problems of ALV. First, an intelligent method is proposed to enhance the performance of a coupled global positioning/inertial navigation system (GPS/INS) for land navigation applications during the GPS signal loss. Our method is based on using an artificial neural network (ANN) to intelligently aid the GPS/INS coupled navigation system in the absence of GPS signals. The proposed enhanced GPS/INS is used in the dynamic environment of a tour of an autonomous van and we provide the results here. GPS/INS+ANN system performance is thus demonstrated with the land trials. Secondly, our work focuses on the identification and enlargement of the stability region of the ALV. In this thesis, the domain of attraction of the ALV is found to be patched by chaotic and regular regions with chaotic boundaries which are extracted using novel technique of cell mapping equipped with measures of fractal dimension and rough sets. All image cells in the cellular state space, with their individual fractal dimension are classified as being members of lower approximation (surely stable), upper approximation (possibly stable) or boundary region using rough set theory. The obtained rough set with fractal dimension as its attribute is used to model the uncertainty of the regular regions. This uncertainty is then smoothed by a reinforcement learning algorithm in order to enlarge regular regions that are used for chassis control, critical in ALV in preventing vibration damages that can harm the payload. Hence, we will make ALV work in the largest safe area in dynamical sense and prevent the vehicle and its cargo.

No description available.

Inertial Measurement Unit (IMU) and Global Positioning System (GPS) receivers are sensors that are widely used for land vehicle navigation. GPS receivers provide position and/or velocity data to any user on the Earth&rsquo / s surface independent of his position. Yet, there are some conditions that the receiver encounters difficulties, such as weather conditions and some blockage problems due to buildings, trees etc. Due to these difficulties, GPS receivers&rsquo / errors increase. On the other hand, IMU works with respect to Newton&rsquo / s laws. Thus, in stark contrast with other navigation sensors (i.e. radar, ultrasonic sensors etc.), it is not corrupted by external signals. Owing to this feature, IMU is used in almost all navigation applications. However, it has some disadvantages such as possible alignment errors, computational errors and instrumentation errors (e.g., bias, scale factor, random noise, nonlinearity etc.). Therefore, a fusion or integration of GPS and IMU provides a more accurate navigation data compared to only GPS or only IMU navigation data. v In this thesis, loosely coupled GPS/IMU integration systems are implemented using feed forward and feedback configurations. The mechanization equations, which convert the IMU navigation data (i.e. acceleration and angular velocity components) with respect to an inertial reference frame to position, velocity and orientation data with respect to any desired frame, are derived for the geographical frame. In other words, the mechanization equations convert the IMU data to the Inertial Navigation System (INS) data. Concerning this conversion, error model of INS is developed using the perturbation of the mechanization equations and adding the IMU&rsquo / s sensor&rsquo / s error model to the perturbed mechanization equation. Based on this error model, a Kalman filter is constructed. Finally, current navigation data is calculated using IMU data with the help of the mechanization equations. GPS receiver supplies external measurement data to Kalman filter. Kalman filter estimates the error of INS using the error mathematical model and current navigation data is updated using Kalman filter error estimates. Within the scope of this study, some real experimental tests are carried out using the software developed as a part of this study. The test results verify that feedback GPS/INS integration is more accurate and reliable than feed forward GPS/INS. In addition, some tests are carried out to observe the results when the GPS receiver&rsquo / s data lost. In these tests also, the feedback GPS/INS integration is observed to have better performance than the feed forward GPS/INS integration.