Sharif Digital Repository

The Gaussian mixture probability hypothesis density (GM-PHD) filter is a closed form approximation of the multi-target Bayes filter which can overcome most multitarget tracking problems. Limited field of view, decreasing cost of cameras, and advances of using multi-camera induce us to use large-scale camera networks. In this paper, a multihuman tracking framework using the PHD filter in a distributed camera network is proposed. Each camera tracks objects locally with PHD filter and a track-after-detect scheme and its estimates of targets are sent to neighboring nodes. Then each camera fuses its local estimates with it's neighbors. The proposed method is evaluated on the public PETS2009... 

Asynchronous data fusion is inevitable in track-to-track fusion for tracking high-speed targets. For low-speed targets, e.g., the movement of clouds, synchronization is insignificant and, depending on the application, may be disregarded. Real-time asynchronous fusion is a demanding task in sensor networks when the sensors are not synchronous in sampling-rate or in sampling-phase. In the method proposed in this paper, an estimator in the fusion center estimates the actual time of the sample with respect to the time-reference of the fusion center upon receiving the data from a sensor. Then, the computer of the fusion center uses predictions to transfer all the received data to the data... 

Concurrent orbit and attitude determination (COAD) plays a key role in reducing the cost of navigation and control subsystem for small satellites. This article is devoted to the problem of the COAD of satellites. A measurement package consisting of three axis magnetometer (TAM) and a sun sensor is shown to be sufficient to estimate the attitude and orbit information. To this end, an autonomous gyro-less COAD algorithm is proposed and implemented through the centralized data fusion of the TAM and the sun sensor. The set of nonlinear-coupled roto-translation dynamics of the satellite is used with a modified unscented Kalman filter (MUKF) to estimate the full satellite states. The MUKF is... 

Multi-sensor tracking is widely used for augmentation of tracking accuracy using data fusion. A basic requirement for such applications is the real time temporal synchronization and spatial registration of two sensory data. In this study a new method for time and space coordination of two tracking sensor measurements has been presented. For spatial registration we used a body coordinate system and then applied the effect of the level arm. The time synchronization was done based on least mean square (LMS) error method. This method was implemented to synchronize the position and orientation of an object using Inertial (1IMU) and Optical (Optotrak) tracking systems. The results of synchronized... 

Efficient combination of MIMO (multiple-input multiple-output) and PCL (passive coherent location) ideas is expected to improve performance of localization schemes. While, in general, adding the number of transmit and receive antennas provides spatial diversity, it is possible to obtain similar effects by using multiple transmitters that are already transmitting standard signals (such as digital TV (DTV) transmission) in the environment (also known as illuminators of opportunity). However, in the case of passive schemes, it is not always possible to ensure that signals of different transmitters are orthogonal to each other. In such cases in which nonorthogonal transmitter signals are to be... 

Human tracking is an essential step in many computer vision-based applications. As single view tracking may not be sufficiently robust and accurate, tracking based on multiple cameras has been widely considered in recent years. This paper presents a distributed human tracking method in a smart camera network and introduces a particle filter design based on Histogram of Oriented Gradients (HOG) and color histogram. The proposed adaptive motion model also estimates the target speed from the history of its latest displacement and improves the robustness of the tracker by decreasing the probability of missing targets. In addition, a distributed data fusion method is proposed which fuses the... 

The problem of gyrocompassing using inertial sensors, i.e., gyros and accelerometers, is addressed. North finding, with an order of accuracy of one arc-min, is not only required for the initial alignment of inertial navigation systems, but also has a critical role to play in the guidance and navigation of ships that navigate for long periods of time. In this work, after extracting the error model of an inertial navigation system and augmenting it with the error model of inertial sensors, a processing algorithm based on the Kalman filter is designed and simulated to process the navigation system velocity error, and to estimate and correct tilt and heading errors along with gyro drifts and... 

Distributed spectrum sensing (DSS) is of great importance in Cognitive Radio, especially under fading or shadowing effects. In order to evaluate the performance of a distributed system, it is commonly compared with the centralized system as an upper performance bound. Now the question is whether or not one can obtain a distributed strategy serving as an upper bound to benchmark any distributed strategy, tighter than that of the centralized scheme. Here, we suggest employing the Neyman-Pearson (NP) fusion rule to achieve an upper bound. Furthermore, the analysis of a randomized fusion rule has been provided, which is a long-existing problem in this field. For this purpose, theoretical... 

An algorithm, by cooperation of multiple flying vehicles, is developed to localize a moving target in the presence of measurement noise and mis-modeling. It works based on jointly sharing information and data fusion by using a recursive Bayesian estimator and a searching guidance law to direct each flying vehicle to a position where the probability of target detection is maximum. To evaluate this algorithm, a high fidelity simulation program with six degrees of freedom dynamics is also developed  

Facial Action Coding System consists of 44 action units (AUs) and more than 7000 combinations. Hidden Markov models (HMMs) classifier has been used successfully to recognize facial action units (AUs) and expressions due to its ability to deal with AU dynamics. However, a separate HMM is necessary for each single AU and each AU combination. Since combinations of AU numbering in thousands, a more efficient method will be needed. In this paper an accurate real-time sequence-based system for representation and recognition of facial AUs is presented. Our system has the following characteristics: 1) employing a mixture of HMMs and neural network, we develop a novel accurate classifier, which can... 

Human motion tracking has many applications in biomedical and industrial services. Low-cost inertial/magnetic sensors are widely used in human motion capture systems to obtain the orientation of the human body segments. In this paper, we have presented a quaternion-based unscented Kalman filter algorithm to fuse inertial/magnetic sensors measurements for tracking human arm movements. In order to have a better estimation of the orientation of the forearm and the upper arm, a constraint equation was developed based on the relative velocity of the elbow joint with respect to the inertial sensors attached to the forearm and the upper arm. Also to compensate for fast body motions, we adapted the... 

Human movement analysis is an important part of biomechanics and rehabilitation, for which many measurement systems are introduced. Among these, wearable devices have substantial biomedical applications, primarily since they can be implemented both in indoor and outdoor applications. In this study, a Trunk Motion System (TMS) using printed Body‐Worn Sensors (BWS) is designed and developed. TMS can measure three‐dimensional (3D) trunk motions, is lightweight, and is a portable and non‐invasive system. After the recognition of sensor locations, twelve BWSs were printed on stretchable clothing with the purpose of measuring the 3D trunk movements. To integrate BWSs data, a neural network data... 

We define model-data interaction (MDI) as a two way process between models and data, in which on one hand data can serve the modeling purpose by supporting model discrimination, parameter refinement, uncertainty analysis, etc., and on the other hand models provide a tool for data fusion, interpretation, interpolation, etc. MDI has many applications in the realm of groundwater and has been the topic of extensive research in the groundwater community for the past several decades. This has led to the development of a multitude of increasingly sophisticated methods. The progress of data acquisition technologies and the evolution of models are continuously changing the landscape of groundwater... 

The interest in wearable systems among the biomedical engineering and clinical community continues to escalate as technical refinements enhance their potential use for both indoor and outdoor applications. For example, an important wearable technology known as a microelectromechanical system (MEMS) is demonstrating promising applications in the area of biomedical engineering. Accordingly, this study was designed to investigate the Sharif-Human Movement Instrumentation System (SHARIF-HMIS), consisting of inertial measurement units (IMUs), stretchable clothing, and a data logger—all of which can be used outside the controlled environment of a laboratory, thus enhancing its overall utility.... 

This work describes an efficient technique to sequentially combine estimates resulting from individual sets of measurements provided by a network of satellites. The prescribed method is especially effective to estimate motion states of an uncooperative space object using range image data. The technique, which is fast and suitable for on-line applications, could also be effective to capture stray objects or those satellites that require periodic servicing. Such missions call for high degree of precision and reliable estimation methods. In fact, the proposed estimation architecture consists of a network of synchronized platforms, i.e., Observer Satellites (OS), each with processing power and... 

With the advent of new positioning services, one can reach more satellites these days to find his position. Using a combination of satellites which belong to different constellations needs some considerations like addressing biases between their time references. Each constellation has progressed to the point that they have enough satellites to provide accurate position separately. According to this fact, we propose to find the position in each constellation and fusion their results in a way that final position has the minimum possible variance instead of combining the constellations in a satellite level and dealing with inter system biases. Experimental studies are conducted based on IGS... 

Human motion tracking has many applications in biomedical and industrial services. Low-cost inertial/magnetic sensors are widely used in human motion capture systems to obtain the orientation of the human body segments. In this paper, we have presented a quaternion-based unscented Kalman filter algorithm to fuse inertial/magnetic sensors measurements for tracking human arm movements. In order to have a better estimation of the orientation of the forearm and the upper arm, a constraint equation was developed based on the relative velocity of the elbow joint with respect to the inertial sensors attached to the forearm and the upper arm. Also to compensate for fast body motions, we adapted the... 

By development of ocean exploration, autonomous vehicles are employed to perform on-water and underwater tasks. Using an extended Kalman filter, Inertial Navigation System/Doppler Velocity Log (INS/DVL) integrated systems are trying to navigate in oceans and underwater environments when Global Positioning System (GPS) signals are not accessible. The dependency of DVL signals on acoustic environments may cause any DVL malfunction due to sea creatures or strong wave-absorbing material. In this paper, an improved version of evolutionary TS-fuzzy (eTS) is proposed in order to predict DVL sensor outputs at DVL malfunction moment, by utilizing an artificial intelligent (AI) aided integrated... 

In this paper, we investigate the problem of detecting and monitoring changes (abnormality) in molecular communication (MC), using the quickest change detection (QCD) schemes. The objective is to watch an environment using a sensor network and make decisions on the time and location of changes based on the received signals from sensors in the fusion center (FC). Such assumptions call for considering spatial and temporal correlations among sensors' transmitting signals. We use the framework of Partially Observable Markov Decision Processes (POMDPs) based on non-homogeneous Markov models. The metric in detection (stopping-time) scenario is to minimize the delay of announcing an abnormality... 

This paper presents a system based on pedestrian dead reckoning (PDR) for localization of networked mobile users, which relies only on sensors embedded in the devices and device- to-device connectivity. The user trajectory is reconstructed by measuring step by step the user displacements. Though step length can be estimated rather accurately, heading evaluation is extremely problematic in indoor environments. Magnetometer is typically used, however measurements are strongly perturbed. To improve the location accuracy, this paper proposes a novel cooperative system to estimate the direction of motion based on a machine learning approach for perturbation detection and filtering, combined with...