Downloads & Free Reading Options - Results

The paper addresses distributed multi-target tracking in the framework of generalized Covariance Intersection (GCI) over multistatic radar system. The proposed method is based on the unlabeled version of generalized labeled multi-Bernoulli (GLMB) family by discarding the labels, referred as generalized multi-Bernoulli (GMB) family. However, it doesn't permit closed form solution for GCI fusion with GMB family. To solve this challenging problem, firstly, we propose an efficient approximation to the GMB family which preserves both the probability hypothesis density (PHD) and cardinality distribution, named as second-order approximation of GMB (SO-GMB) density. Then, we derive explicit expression for the GCI fusion with SO-GMB density. Finally, we compare the first-order approximation of GMB (FO-GMB) density with SO-GMB density in two scenarios and make a concrete analysis of the advantages of the second-order approximation. Simulation results are presented to verify the proposed approach.

An imaging system which applies knowledge-based technology to supervise and control both sensor hardware and computation in the imaging system is described. It includes the development of an imaging system breadboard which brings together into one system work that we and others have pursued for LaRC for several years. The goal is to combine Digital Signal Processing (DSP) with Knowledge-Based Processing and also include Neural Net processing. The system is considered a smart camera. Imagine that there is a microgravity experiment on-board Space Station Freedom with a high frame rate, high resolution camera. All the data cannot possibly be acquired from a laboratory on Earth. In fact, only a small fraction of the data will be received. Again, imagine being responsible for some experiments on Mars with the Mars Rover: the data rate is a few kilobits per second for data from several sensors and instruments. Would it not be preferable to have a smart system which would have some human knowledge and yet follow some instructions and attempt to make the best use of the limited bandwidth for transmission. The system concept, current status of the breadboard system and some recent experiments at the Mars-like Amboy Lava Fields in California are discussed.

In the Command and Control mission, new technologies such as 'sensor fusion' are designed to help reduce operator workload and increase situational awareness. This thesis explored the tracking characteristics of diverse sensors and sources of data and their contributions to a fused tactical picture. The fundamental building blocks of any sensor fusion algorithm are the tracking algorithm associated with each of the sensors on the sensor platform. In support of this study the MATLAB program 'fusim' was written to provide acquisition managers a tool for evaluating tracking and sensor fusion algorithms. The fusim program gives the user flexibility in selecting: sensor platforms up to four sensors associated with that platform, the target types the problem orientation and the tracking algorithms to be used with the sensors. The fusim program was used to compare tracking algorithms in a multiple sensor/multiple target environment. Specifically, the Probabilistic Data Association Filter the Interacting Multiple Models Filter the Kalman Filter and the Constant Gain Kalman Filter were evaluated against multiple maneuvering non- maneuvering, and fixed targets. It is recommended that this study be continued to evaluate advanced tracking and data association techniques to expand the program to allow attribute tracking and identification and to study the Human-Machine Interface aspects of sensor fusion.

This study formulated a distributed agent system for the implementation of an optimal fusion data strategy in a battlefield domain. The approach followed in the study is unique because it doesn't attempt to transform the data to a common representation, but rather, it uses a Multiple Agent Hybrid Estimation Architectural (MAHEA) framework. The optimal state estimate is obtained dynamically by the interation between the individual agents in the architecture over time. Each agent generate it's own state estimate. Each agent's estimate is improved over time by the heterogeneous data flowing from the other agents. The agents' basic block is a variation optimization algorithm and synchronization between agents is based on Noether's invariance principles. The architecture of the agents includes an adaptation mechanism for improving the estimates over time and also a planning element for generating the variational criterion online. The present study improves the formulation of the inference based procedure used for computing the optimal estimate used in an earlier version of the MAHEA architecture. The report illustrates the scheme and architecture with a simple example.

As the applications for using small Unmanned Aircraft Systems (sUAS) beyond visual line of sight (BVLOS) continue to grow in the coming years, it is imperative that intelligent sensor fusion techniques be explored. In BVLOS scenarios the vehicle position must accurately be tracked over time to ensure no two vehicles collide with one another, no vehicle crashes into surrounding structures, and to identify off-nominal scenarios. Therefore, in this study an intelligent systems approach is used to estimate the position of sUAS given a variety of sensor platforms, including, GPS, radar, and on-board detection hardware. Common research challenges include, asynchronous sensor rates and sensor reliability. In an effort to realize these challenges, techniques such as a Maximum a Posteriori estimation and a Fuzzy Logic based sensor confidence determination are used.

Different sensors exploit different regions of the electromagnetic spectrum; therefore a multi-sensor image fusion system can take full advantage of the complementary capabilities of individual sensors in the suit; to produce information that cannot be obtained by viewing the images separately. In this thesis a framework for the multi resolution fusion of the night vision devices and thermal infrared imagery is presented. It encompasses a wavelet-based approach that supports both pixel-level and region-based fusion and aims to maximize scene content by incorporating spectral information from both the source images. In pixel-level fusion source images are decomposed into different scales and salient directional features are extracted and selectively fused together by comparing the corresponding wavelet coefficients. To increase the degree of subject relevance in the fusion process a region-based approach which uses a multiresolution segmentation algorithm to partition the image domain at different scales is proposed. The regions characteristics are then determined and used to guide the fusion process. The experimental results obtained demonstrate the feasibility of the approach. Potential applications of this development include improvements in night piloting (navigation and target discrimination) law enforcement etc.

In the Command and Control mission, new technologies such as 'sensor fusion' are designed to help reduce operator workload and increase situational awareness. This thesis explored the tracking characteristics of diverse sensors and sources of data and their contributions to a fused tactical picture. The fundamental building blocks of any sensor fusion algorithm are the tracking algorithm associated with each of the sensors on the sensor platform. In support of this study the MATLAB program 'fusim' was written to provide acquisition managers a tool for evaluating tracking and sensor fusion algorithms. The fusim program gives the user flexibility in selecting: sensor platforms up to four sensors associated with that platform, the target types the problem orientation and the tracking algorithms to be used with the sensors. The fusim program was used to compare tracking algorithms in a multiple sensor/multiple target environment. Specifically, the Probabilistic Data Association Filter the Interacting Multiple Models Filter the Kalman Filter and the Constant Gain Kalman Filter were evaluated against multiple maneuvering non- maneuvering, and fixed targets. It is recommended that this study be continued to evaluate advanced tracking and data association techniques to expand the program to allow attribute tracking and identification and to study the Human-Machine Interface aspects of sensor fusion.

In the Command and Control mission, new technologies such as 'sensor fusion' are designed to help reduce operator workload and increase situational awareness. This thesis explored the tracking characteristics of diverse sensors and sources of data and their contributions to a fused tactical picture. The fundamental building blocks of any sensor fusion algorithm are the tracking algorithm associated with each of the sensors on the sensor platform. In support of this study the MATLAB program 'fusim' was written to provide acquisition managers a tool for evaluating tracking and sensor fusion algorithms. The fusim program gives the user flexibility in selecting: sensor platforms up to four sensors associated with that platform, the target types the problem orientation and the tracking algorithms to be used with the sensors. The fusim program was used to compare tracking algorithms in a multiple sensor/multiple target environment. Specifically, the Probabilistic Data Association Filter the Interacting Multiple Models Filter the Kalman Filter and the Constant Gain Kalman Filter were evaluated against multiple maneuvering non- maneuvering, and fixed targets. It is recommended that this study be continued to evaluate advanced tracking and data association techniques to expand the program to allow attribute tracking and identification and to study the Human-Machine Interface aspects of sensor fusion.

Title from cover

A study on filtering aspects of airborne wind energy generators is presented. This class of renewable energy systems aims to convert the aerodynamic forces generated by tethered wings, flying in closed paths transverse to the wind flow, into electricity. The accurate reconstruction of the wing's position, velocity and heading is of fundamental importance for the automatic control of these kinds of systems. The difficulty of the estimation problem arises from the nonlinear dynamics, wide speed range, large accelerations and fast changes of direction that the wing experiences during operation. It is shown that the overall nonlinear system has a specific structure allowing its partitioning into sub-systems, hence leading to a series of simpler filtering problems. Different sensor setups are then considered, and the related sensor fusion algorithms are presented. The results of experimental tests carried out with a small-scale prototype and wings of different sizes are discussed. The designed filtering algorithms rely purely on kinematic laws, hence they are independent from features like wing area, aerodynamic efficiency, mass, etc. Therefore, the presented results are representative also of systems with larger size and different wing design, different number of tethers and/or rigid wings.

A highly-scalable, Bayesian approach to the problem of performing multi-source data fusion and target tracking in decentralized sensor networks is presented. Previous applications of decentralized data fusion have generally been restricted to uni-modal/uni-source sensor networks using Gaussian based approaches, such as the Kalman or information filter. However, with recent interest to employ complex, multimodal/ multi-source sensors which potentially exhibit observation and/or process non-linearities along with non-Gaussian distributions, the need to develop a more generalized and scalable method of decentralized data fusion is required. The probabilistic approach featured in this work provides the ability to seamlessly integrate and efficiently fuse multi-source sensor data in the absence of any linearity and/or normality constraints. The proposed architecture is fully decentralized and provides a methodology that scales extremely well to any growth in the number of targets or region of coverage. This paper will illustrate that our multi-source data fusion architecture is capable of providing high-precision tracking performance in complex, non-linear/non-Gaussian operating environments. In addition, we will also show that our architecture provides an unprecedented scaling capability for decentralized sensor networks as compared to similar architectures which communicate information using particle data, Gaussian mixture models or Parzen density estimators.

This paper describes the authors' efforts to develop a pedigree ontology for level-one sensor fusion. This work was performed in the context of naval operations but the concepts employed are applicable to any domain involving sensor fusion. The ontology that has been developed is formally represented using OWL, the Web Ontology Language used in defining ontologies for the Semantic Web. One advantage of using OWL is that it has a formal semantics and there is a growing number of formal systems for processing and reasoning about OWL-based documents. The paper will describe the pedigree ontology in detail along with the motivation for various design decisions. An example will be given of the ontology's use in conjunction with OTH-T GOLD track data; in a simulated scenario pedigree information permits the informed selection of preferred vessel tracks. The paper concludes with a discussion of open issues and future directions.

Schemes for quantization and fusion in multi-sensor systems used for discriminating between two sequences of dependent observations are introduced and analyzed. The observation sequences of each sensor under the two hypotheses are arbitrary stationary dependent sequences that can not be modeled as signal in additive noise; the objective of the fusion center is to discriminate between the two hypotheses. These observation models are well motivated by practical multi-sensor target discrimination problems. Two cases are considered: in the first, the observation sequences of the sensors are individually dependent but jointly mutually independent; in the second case, the observation sequences are dependent across both time and sensors. The dependence in the observations across time and/or sensors is modeled by m-dependent, phi mixing, or rho-mixing processes. The following four quantization/fusion schemes are considered: (a) forming test statistics at the sensors by passing the observations through memoryless non-linearities, summing them up, and fusing these test statistics without previous quantization; (b) quantizing uniformally (with equidistant breakpoints) each sensor observation and then fusing; (c) quantizing optimally each sensor observation and then fusing; and (d) using the sensor test statistic of (a) to make binary decisions and then fusing the binary decisions. To guarantee high-quality performance, a common large sample size is employed by each sensor and an asymptotic analysis is pursued. Design criteria are developed from the Bayesian cost of the fusion center for deriving the optimal memoryless nonlinearities of the sensor test statistics and the sensor quantizer parameters (quantization levels and breakpoints). These design criteria are shown to involve an extension of the generalized signal-to-noise ratio used in single-sensor detection and quantization.

Condition monitoring is a predictive maintenance procedure, which is used to provide warning and predict faults of a system at an early stage, so catastrophic breakdown of the system can be avoided. An efficient condition monitoring method can reduce the cost of breakdown maintenance, increase the working life of system, and reduce the spare part inventory. Condition monitoring system acquires raw data of different system parameters using different sensors. Different signal processing techniques are implemented on the raw data to extract vital clues and estimate the probable faults in the machine. This paper gives an overview of condition monitoring system of induction motor and gives an overview of implementation of multi sensor data fusion technique to improve the efficiency of condition monitoring system.

Multi-modal perception systems based on multiple sensors and computer vision technology are the technical prerequisite for the stable operation of intelligent driving. However, in the face of abnormal sensor feedback, such as interference with the visual system caused by extreme weather or blurring of sensors caused by complex road conditions, autonomous driving systems face reliability challenges. Such sensor anomalies are likely to cause system logic conflicts, leading to fatal consequences, such as the inability to make accurate judgments quickly when driving at high speeds. This survey comprehensively reviews the impact of abnormal environments on CV-based image systems and LiDAR-based sensors. This review conducts an in-depth analysis of multi-modal fusion technology and its contribution to the reliability of autonomous driving systems in extreme environments based on some advanced work in recent years. Finally, the challenges faced by autonomous driving systems, unresolved issues, and potential future research directions are emphasized.

Tomographic Data Fusion with CFD Simulations Associated with a Planar Sensor 作者： LIU J. 1 LIU S. 1 SUN S. 1 ZHOU W. 1 I.H.I Schlaberg 1 WANG M. 1 YAN Y. 1 作者单位： 1. North China Electric Power University, 2 Beinong Road, Changping District, Beijing 102206, China 提交时间： 2017-03-31 摘要: Tomographic techniques have great abilities to interrogate the combustion processes, especially when it is com- bined with the physical models of the combustion itself. In this study, a data fusion algorithm is developed to in- vestigate the flame distribution of a swirl-induced environmental (EV) burner, a new type of burner for low NOx combustion. An electric capacitance tomography (ECT) system is used to acquire 3D flame images and computa- tional fluid dynamics (CFD) is applied to calculate an initial distribution of the temperature profile for the EV burner. Experiments were also carried out to visualize flames at a series of locations above the burner. While the ECT images essentially agree with the CFD temperature distribution, discrepancies exist at a certain height. When data fusion is applied, the discrepancy is visibly reduced and the ECT images are improved. The methods used in this study can lead to a new route where combustion visualization can be much improved and applied to clean energy conversion and new burner development. Process tomography data fusion EV burner flame 分类： 动力与电气工程 >> 工程热物理学 引用： ChinaXiv:201703.01051 (或此版本 ChinaXiv:201703.01051V1 ) doi:10.12074/201703.01051 CSTR:32003.36.ChinaXiv.201703.01051.V1 推荐引用方式： LIU J.,LIU S.,SUN S.,ZHOU W.,I.H.I Schlaberg,WANG M.,YAN Y..(2017).Tomographic Data Fusion with CFD Simulations Associated with a Planar Sensor.热科学学报.[ChinaXiv:201703.01051] 版本历史 [V1] 2017-03-31 10:35:19 ChinaXiv:201703.01051V1 下载全文

The Dempster-Shafer Theory, a generalization of the Bayesian theory, is based on the idea of belief and as such can handle ignorance. When all of the required information is available, many data fusion methods provide a solid approach. Yet, most do not have a good way of dealing with ignorance. In the absence of information, these methods must then make assumptions about the sensor data. However, the real data may not fit well within the assumed model. Consequently, the results are often unsatisfactory and inconsistent. The Dempster-Shafer Theory is not hindered by incomplete models or by the lack of prior information. Evidence is assigned based solely on what is known, and nothing is assumed. Hence, it can provide a fast and accurate means for multi-sensor fusion with ignorance. In this research, we apply the Dempster-Shafer Theory in target tracking and in gait analysis. We also discuss the Dempster-Shafer framework for fusing data from a Global Positioning System (GPS) and an Inertial Measurement Unit (IMU) sensor unit for precise local navigation. Within this application, we present solutions where GPS outages occur.

Mulitsensor data fusion (MSDF) has been researched for decades yet programs relying on it to provide a situational, or threat, assessment continue to be less than successful. In order to alleviate the too-much-information, too-few-analysts issue, a better approach must be determined. A survey of recent and current data fusions programs was conducted along with a literature review on how different organizations handle a fusion-based assessment. Key points found in this study were used to develop an adaption of models that can be used to provide an improved assessment while simplifying the process needed to get there.

The overall objective of this research is to improve and enhance MUSSLE (Multiple Sensor Statistical Likelihood Estimator), the Multi-Sensor Integration algorithm under development for the E-3 AWACS platform. Modifications to the software completed to date include: enhanced three dimensional report processing for the Contact-to-Track Association module, MATCH; restructuring of the MUSSLE database so that ambiguous correlations are maintained in a cluster; improvements to the throughput speed of the algorithm; improvements to the passive tracking module; and improvements to the responses to operator switch actions. In addition to the code modifications, we produced a throughput and memory requirements analysis of the MUSSLE program.

The overall objective of this research is to improve and enhance MUSSLE (Multiple Sensor Statistical Likelihood Estimator), the Multi-Sensor Integration algorithm under development for the E-3 AWACS platform. Modifications to the software completed to date include: enhanced three dimensional report processing for the Contact-to-Track Association module, MATCH; restructuring of the MUSSLE database so that ambiguous correlations are maintained in a cluster; improvements to the throughput speed of the algorithm; improvements to the passive tracking module; and improvements to the responses to operator switch actions. In addition to the code modifications, we produced a throughput and memory requirements analysis of the MUSSLE program.

This report addresses the problem of data fusion and sensor management based on a synergistic use of the information provided by multiple sensors. In the first studied case, we deal with the problem of measurement strategy computation for a passive receiver. The basic problem is then to compute an optimal policy, during a specified observation time interval so that a prediction accuracy is optimized. It is shown that the optimal measurement policy can be precomputed before the measurements actually occur. The second case descibes research work on the selection of a strategy of measurements for an active and a passive systems. The approach is based on selecting at each instant of time, a set of measurements provided by one or more sensors. Each sensor measurement has an associated cost. The basic problem is then to select an optimal measurement policy, during a specified receding horizon observation interval, so that a weighted combination of prediction accuracy and observation cost is optimized.

Bayesian techniques have been employed to combine, or fuse, reports from multiple target identification (ID) sensors. Recent work has shown that a 'cost' criterion can be used to make optimum target declaration decisions from the Bayesian representations of the ID sensors. However, the 'cost' method, as previously formulated, only works when choosing between two types of target declarations (e.g., friend vs. hostile). This paper provides an alternative formulation to the cost methodology previously developed for Bayesian analysis of ID sensor fusion. This alternative formulation is then used to develop a formulation that works for multiple types of targets. Therefore, this alternative formulation will allow analysts and military commanders to include neutral targets in the ID fusion process. Indeed, the technique also enables optimum ID sensor fusion when considering multiple additional types of targets (e.g., coalition force, hostile fighters, hostile bombers, and hostile cruise missiles).

This thesis serves as an exploration that takes the sensors within a cell phone beyond the current state of recognition activities. Current state of the art sensor recognition processes tend to focus on recognizing user activity. Utilizing the same sensors available for user activity classi cation, this thesis validates the ability to gather data about entities separate from the user carrying the smart phone. With the ability to sense entities, the ability to recognize and classify a multitude of items, situations, and phenomena opens a new realm of possibilities for how devices perceive and react to their environment.

In this paper, we explore the use of the Dezert-Smarandache Theory (DSmT) for seismic and acoustic sensor fusion. The seismic/acoustic data is noisy which leads to classification errors and conflicts in declarations. DSmT affords the redistribution of masses when there is a conflict.

This paper describes a distributed sensor-data-fusion paradigm and theory based on a previously developed theory to model sensors as highly autonomous units. Generic procedures are defined to reason and make decisions at the qualitative level. This facilitates distribution of intelligence ( code and hardware) to the sensor level and peer-to-peer communication among sensors, controllers, and other devices in the system.

The fire safety system is an important component that controls material and immaterial losses. Fire disasters are generally indicated by the appearance of excess smoke and changes in temperature, pressure, and changes in other parameters in the environment. Conventional smoke sensors are limited in reading parameter changes around their environment, making them less effective in early fire detection. This study aims to design a smoke detection system as an early fire detection system, using sensor fusion based on deep learning using the recursive feature elimination method with cross-validation (RFECV) using a random forest classifier used to select optimal parameters from public datasets as the basis for determining the sensor to be used. Based on the RFECV optimal feature, a deep learning algorithm was performed and obtained an accuracy of 0.99, a precision of 0.99, a recall of 1.00, and an F1 score of 0.99, with a latency time of 34.02 μs, which is 71.76% times faster than the original model.

In this paper, we explore the use of DSMT for seismic and acoustic sensor fusion. The seismic/acoustic data is noisy which leads to classification errors and conflicts in declarations. DSmT affords the redistribution of masses when there is a conflict. The goal of this paper is to present an application and comparison on DSMT with other classifier methods to include the support vector machine(SVM) and Dempster-Shafer methods. 

A new multi-sensor fusion approach for integrated ship motion perception in inland waterways 作者： Yong Wu 1 Xiuming Chu 1,2,3 Lei Deng 4 Jinyu Lei 2 Wei He 2 Grzegorz Królczyk 5 Zhixiong Li 5 作者单位： 1. National Engineering Research Center for Water Transport Safety, Wuhan University of Technology, Wuhan 430063, China 2. College of Physics and Electronic Information Engineering, Minjiang University, Fuzhou 350108, China 3. Fujian Engineering Research Center of Safety Control for Ship Intelligent Navigation, Minjiang University, Fuzhou 350108, China 4. School of Electrical, Computer & Telecommunications Engineering, University of Wollongong, New South Wales 2522, Australia 5. Faculty of Mechanical Engineering, Opole University of Technology, Opole 45-758, Poland 通讯作者： Wei He Email:[email protected] 提交时间： 2024-03-31 21:04:21 摘要: The ship motion perception approaches mainly use maritime radar, Automatic Identification System (AIS) and cameras. However, using either of these approaches alone may result in information inconsistency and insufficient data accuracy. Therefore, a multi-sensor fusion perception system is proposed in this study to monitor ship motion in inland waterways. Firstly, a hardware platform of multi-sensor fusion ship motion perception system composed of maritime radar, AIS, cameras and other accessories is constructed. Secondly, by utilizing the target detection and tracking algorithms, track association algorithms, the ship motion data collected from the three sensors are integrated. Finally, the performance of the ship motion perception system is verified by field experiments in day and night. The experimental results indicate that the integrated ship motion perception system with multiple sensors is able to improve the information consistency and data accuracy of ship motion apparently in inland waterway compared to other perception systems. motion perception maritime radar AIS target detection and tracking information fusion 来自： Zhixiong Li 分类： 交通运输工程 >> 水路运输 投稿状态： 已被期刊接收 引用： ChinaXiv:202404.00003 (或此版本 ChinaXiv:202404.00003V1 ) DOI:10.12074/202404.00003V1 CSTR:32003.36.ChinaXiv.202404.00003.V1 推荐引用方式： Yong Wu,Xiuming Chu,Lei Deng,Jinyu Lei,Wei He,Grzegorz Królczyk,Zhixiong Li.(2024).A new multi-sensor fusion approach for integrated ship motion perception in inland waterways.中国科学院科技论文预发布平台.[ChinaXiv:202404.00003] 版本历史 [V1] 2024-03-31 21:04:21 ChinaXiv:202404.00003V1 下载全文

A high-level data fusion system that adopts Bayesian statistics involving weights of evidence modelling is described to combine disparate information from airborne digital data such as a digital surface model (DSM), colour (ADS40), thermal infrared (TABI) and hyperspectral images (AISA) of different period. To show the efficacy of the system, an analysis of subtle change-detection is demonstrated. The data fusion system is capable of detecting changes in man-made features automatically in a densely populated area where there are no much prior information. Multi-class segmented images are obtained from the data captured by the four airborne remote sensing sensors. The system performs data fusion modelling by using binary images of each theme classes and thus a total of 40 binary patterns are obtained. Through Bayesian methods, involving weights of evidence modelling, all the binary images are analysed and finally four binary patterns (indicator images) are identified automatically as significant for the change-detection application. A weighted index overlay model available in the system, combines these four patterns. Data fusion by the weights of evidence modelling is found to be candid and unequivocal for predicting newly transformed locations. Results of Bayesian method are accurate as the weights are based on statistical analysis. Changes in features such as colour of roof, parking areas, openland areas, newly built structures, and presence or absence of vehicle is automatically extracted by using the high-level data fusion approach. The final predictor image showing the probability of change-detected areas in a high-dense city of Japan is generated robustly. The fused image products can be used to update maps for digital cartographic applications. ©2005 Microsoft Corporation. All rights reserved.

By using emittance and reflectance radiation models, the effects of angle of observation, polarization, and spectral content are analyzed to characterize the geometrical and physical properties--reflectivity, emissivity, orientation, dielectric properties, and roughness--of a sensed surface. Based on this analysis, the use of microwave, infrared, and optical sensing is investigated to assure the perception of surfaces on a typical lunar outpost. Also, the concept of employing several sensors on a lunar outpost is explored. An approach for efficient hardware implementation of the fused sensor systems is discussed.

Accurate Automation Corporation (AAC) has developed a novel neural network-based sensor fusion system, inspired by the architecture of biological sensor fusion systems. The project performed research into the nature by which information from multiple sensors is fused by the central nervous system and developed a biological model for the process. Based upon this model we developed a system which fuses two or more sensor signals to generate a fused signal with an improved confidence of target existence and position. The system includes gain, control and fusion units, and also include an integration unit. The integration unit receives signals generated by two or more sensors, and generates integrated signals based on the sensor signals. The integration unit performs temporal and weighted spatial integration of the sensor signals, to generate respective sets of integrated signals supplied to the gain control and fusion units. The gain control unit uses a preprogrammed function to map the integrated signals to an output signal that is scaled to generate a gain signal supplied to the fusion unit. The fusion unit uses a preprogrammed function to map its received integrated signals and the gain signal, to a fused signal that is the output of the system. The weighted spatial integration increases the fused signal's sensitivity to near detections and suppresses response to detections relatively distant in space and time, from a detection of interest. The gain control and fusion functions likewise suppress the fused signal's response to low-level signals, but enhances response to high-level signals. In addition, the gain signal is generated from signals integrated over broad limits so that, if a detection occurred near in space or time to a detection of interest, the gain signal will cause the fused signal to be more sensitive to the level of the detection of interest.

This publication reports the unclassified papers presented at a specialists' meeting held by the Sensor and Propagation Panel at its Fall 1997 meeting. The topics covered included: (1) Applications of multiple sensors and data fusion; (2) techniques and methods; (3) Sensor data networks and management techniques; and (4) Validation studies, experiments, technologies.

This publication reports the unclassified papers presented at a specialists' meeting held by the Sensor and Propagation Panel at its Fall 1997 meeting. The topics covered included: (1) Applications of multiple sensors and data fusion; (2) techniques and methods; (3) Sensor data networks and management techniques; and (4) Validation studies, experiments, technologies.

This paper presents a novel statistical information fusion method to integrate multiple-view sensor data in multi-object tracking applications. The proposed method overcomes the drawbacks of the commonly used Generalized Covariance Intersection method, which considers constant weights allocated for sensors. Our method is based on enhancing the Generalized Covariance Intersection with adaptive weights that are automatically tuned based on the amount of information carried by the measurements from each sensor. To quantify information content, Cauchy-Schwarz divergence is used. Another distinguished characteristic of our method lies in the usage of the Labeled Multi-Bernoulli filter for multi-object tracking, in which the weight of each sensor can be separately adapted for each Bernoulli component of the filter. The results of numerical experiments show that our proposed method can successfully integrate information provided by multiple sensors with different fields of view. In such scenarios, our method significantly outperforms the state of art in terms of inclusion of all existing objects and tracking accuracy.

This report contains a summary of the results of the first two years, and the results obtained during the third year of research in the mathematical problems associated with the analysis and design of Air Force sensor correlation and data fusion systems. These systems play a vital role in the command and control process, but presently there exists no systematic and quantitative methodology for their analysis and design. In the first year of research ALPHATECH investigated an important subproblem: the distributed detection problem associated with determining the presence or absence of targets from a collection of distributed sensors. In the second year of research, ALPHATECH obtained novel exact expressions for the probability density function of the local log likelihood ratios used in distributed detection problems, and has used these expressions to generate an extensive set of design curves. In the third year of research, ALPHATECH investigated another important class of problems, namely, sequential distributed detection problems. Three such problems were formulated: the infinite horizon decentralized Wald problem; a decentralized Quickset Detection problem; and a sequential distributed detection problem with communication and ordered stopping times. The qualitative properties of the optimal solution for the decentralized Wald and the decentralized Wald problem and the sequential distributed detection problem with communication were obtained. (Author)

Distributed multi-sensor fusion has been widely used in military and civilian applications. In the statistical sensor fusion domain, the design of an optimal fusion processor usually requires the joint statistics of the local sensor outputs. When accurate joint statistical knowledge is not readily available, popular solutions are either to estimate the joint statistics from training data or to simply assume independence of the data. Although it is well known that a fusion solution constructed using empirical data or simplified assumptions often cannot reach the optimal performance, little research has been focused on analyzing the performance difference. This paper presents a systematic analysis of distributed sensor fusion performance in an uncertain operating environment using a Bayesian likelihood ratio fusion model. For the problem where joint statistics of the local sensor outputs cannot be obtained accurately, the sub-optimal fusion processor is assumed to have an estimated correlation coefficient and its performance difference from the optimal scenario is derived analytically using a Gaussian model. We use the detectability index, which fully characterizes the receiver operating characteristic (ROC) curve for the Gaussian model, as the performance metric to compare the optimal and suboptimal cases. The ratio of detectability indices for the sub-optimal and optimal cases is derived as a function of the true correlation coefficient, the estimated value, and the performance difference between individual local sensors. We prove that the closer the individual local sensor performances, the less vulnerable the fusion performance is to a mismatched estimation of the correlation coefficient. Furthermore, we show that for the special case where all local sensors have the same performance, the optimal fusion performance is always achieved regardless of the estimation deviation from the true correlation coefficient.

This report summarizes our accomplishments during the most recent period of support under this grant. Our research covers several interrelated areas: (a) statistical modeling methods for complex phenomena using multiresolution, hierarchical, and relational structures; (b)sensor fusion for complex space-time phenomena and activities; (c) development of statistical models for shapes and their use in robust methods for shape estimation and recognition; and (d) methods for blending physics and statistical learning in image reconstruction, feature extraction, and fusion. Our research blends methods from several fields-statistics and probability, signal and image processing, mathematical physics, scientific computing, statistical learning theory, and differential geometry-to produce new approaches to emerging and challenging problems in signal and image processing, and each aspect of our program contains both fundamental research in mathematical sciences and important applications of direct relevance to Air Force missions. In particular, our research is relevant to automatic target recognition based on synthetic aperture radar and laser radar imagery; wide-area surveillance and information preparation of the battlefield; global awareness and higher-level fusion for situational assessment; and fusion of multiple heterogeneous sensors. In all of these areas we have contacts and interactions with AFRL staff and with industry involved in Air Force programs.

The development of intelligent assistive technologies has become increasingly essential in enhancing the mobility and independence of individuals with physical disabilities. This research presents a smart wheelchair system that integrates Internet of Things (IoT) and Artificial Intelligence (AI) with a multi-sensor fusion approach to enable efficient and autonomous navigation. Traditional wheelchairs often require manual operation, which can be challenging for users with severe mobility impairments. The proposed system addresses this limitation by incorporating an array of sensors—ultrasonic sensors, infrared (IR), gyroscope, accelerometer, and GPS—whose data are fused using a Kalman filter algorithm to ensure accurate real-time decision-making for obstacle detection, localization, and path planning. The AI module, built using lightweight deep learning models, enables intelligent environment interpretation and user interaction through voice and gesture recognition. The IoT framework allows for seamless data transmission to a cloud-based monitoring system, enabling caregivers and healthcare professionals to track location, system status, and user safety in real time. Additionally, a mobile application interface enhances user control and connectivity, further improving the overall user experience. Simulation and prototype testing demonstrate the system’s capability to navigate complex indoor and outdoor environments with over 92% obstacle avoidance accuracy and minimal response latency. The integration of sensor fusion and AI significantly improves navigation precision compared to single-sensor models. This project not only showcases the potential of emerging technologies in assistive devices but also offers a scalable solution adaptable to individual user needs. Future work will focus on optimizing energy consumption, improving AI adaptability through reinforcement learning, and conducting extended real-world trials. The proposed smart wheelchair system represents a meaningful step toward empowering individuals with mobility impairments, providing them with increased autonomy, safety, and a higher quality of life through intelligent technology integration.

Crop maps are essential tools for creating crop inventories, forecasting yields, and guiding the use of efficient farm management techniques. These maps must be created at highly exact scales, necessitating difficult, costly, and time-consuming fieldwork. Deep learning algorithms have now significantly enhanced outcomes when using data in the geographical and temporal dimensions, which are essential for agricultural research. The simultaneous availability of Sentinel-1 (synthetic aperture radar) and Sentinel-2 (optical) data provides an excellent chance to combine them. Sentinel 1 and Sentinel 2 data sets were collected for the Cape Town, South Africa, region. With the use of these datasets, we use the fusion technique, particularly the layer-level fusion strategy, one of the three fusion procedures (input level, layer level, and decision level). Also, we will compare the results before and after the fusion and discuss the recommended method for converting from a multilayer perceptron decoder to a semi-supervised decoder architecture. The total testing accuracy produced by the Ada-Match semi-supervised decoder approach was 80.3%. We conduct studies to demonstrate that our methodology not only outperforms prior state-of-the-art approaches in terms of precision but also significantly decreases processing time and memory requirements

Crop maps are essential tools for creating crop inventories, forecasting yields, and guiding the use of efficient farm management techniques. These maps must be created at highly exact scales, necessitating difficult, costly, and time-consuming fieldwork. Deep learning algorithms have now significantly enhanced outcomes when using data in the geographical and temporal dimensions, which are essential for agricultural research. The simultaneous availability of Sentinel-1 (synthetic aperture radar) and Sentinel-2 (optical) data provides an excellent chance to combine them. Sentinel 1 and Sentinel 2 data sets were collected for the Cape Town, South Africa, region. With the use of these datasets, we use the fusion technique, particularly the layer-level fusion strategy, one of the three fusion procedures (input level, layer level, and decision level). Also, we will compare the results before and after the fusion and discuss the recommended method for converting from a multilayer perceptron decoder to a semi-supervised decoder architecture. The total testing accuracy produced by the Ada-Match semi-supervised decoder approach was 80.3%. We conduct studies to demonstrate that our methodology not only outperforms prior state-of-the-art approaches in terms of precision but also significantly decreases processing time and memory requirements

Crop maps are essential tools for creating crop inventories, forecasting yields, and guiding the use of efficient farm management techniques. These maps must be created at highly exact scales, necessitating difficult, costly, and time-consuming fieldwork. Deep learning algorithms have now significantly enhanced outcomes when using data in the geographical and temporal dimensions, which are essential for agricultural research. The simultaneous availability of Sentinel-1 (synthetic aperture radar) and Sentinel-2 (optical) data provides an excellent chance to combine them. Sentinel 1 and Sentinel 2 data sets were collected for the Cape Town, South Africa, region. With the use of these datasets, we use the fusion technique, particularly the layer-level fusion strategy, one of the three fusion procedures (input level, layer level, and decision level). Also, we will compare the results before and after the fusion and discuss the recommended method for converting from a multilayer perceptron decoder to a semi-supervised decoder architecture. The total testing accuracy produced by the Ada-Match semi-supervised decoder approach was 80.3%. We conduct studies to demonstrate that our methodology not only outperforms prior state-of-the-art approaches in terms of precision but also significantly decreases processing time and memory requirements

Crop maps are essential tools for creating crop inventories, forecasting yields, and guiding the use of efficient farm management techniques. These maps must be created at highly exact scales, necessitating difficult, costly, and time-consuming fieldwork. Deep learning algorithms have now significantly enhanced outcomes when using data in the geographical and temporal dimensions, which are essential for agricultural research. The simultaneous availability of Sentinel-1 (synthetic aperture radar) and Sentinel-2 (optical) data provides an excellent chance to combine them. Sentinel 1 and Sentinel 2 data sets were collected for the Cape Town, South Africa, region. With the use of these datasets, we use the fusion technique, particularly the layer-level fusion strategy, one of the three fusion procedures (input level, layer level, and decision level). Also, we will compare the results before and after the fusion and discuss the recommended method for converting from a multilayer perceptron decoder to a semi-supervised decoder architecture. The total testing accuracy produced by the Ada-Match semi-supervised decoder approach was 80.3%. We conduct studies to demonstrate that our methodology not only outperforms prior state-of-the-art approaches in terms of precision but also significantly decreases processing time and memory requirements

This article is from Sensors (Basel, Switzerland) , volume 13 . Abstract Fault diagnosis (FD) and data fusion (DF) technologies implemented in the LabVIEW program were used for a ruthenium dioxide pH sensor array. The purpose of the fault diagnosis and data fusion technologies is to increase the reliability of measured data. Data fusion is a very useful statistical method used for sensor arrays in many fields. Fault diagnosis is used to avoid sensor faults and to measure errors in the electrochemical measurement system, therefore, in this study, we use fault diagnosis to remove any faulty sensors in advance, and then proceed with data fusion in the sensor array. The average, self-adaptive and coefficient of variance data fusion methods are used in this study. The pH electrode is fabricated with ruthenium dioxide (RuO2) sensing membrane using a sputtering system to deposit it onto a silicon substrate, and eight RuO2 pH electrodes are fabricated to form a sensor array for this study.

This article is from Sensors (Basel, Switzerland) , volume 13 . Abstract Fault diagnosis (FD) and data fusion (DF) technologies implemented in the LabVIEW program were used for a ruthenium dioxide pH sensor array. The purpose of the fault diagnosis and data fusion technologies is to increase the reliability of measured data. Data fusion is a very useful statistical method used for sensor arrays in many fields. Fault diagnosis is used to avoid sensor faults and to measure errors in the electrochemical measurement system, therefore, in this study, we use fault diagnosis to remove any faulty sensors in advance, and then proceed with data fusion in the sensor array. The average, self-adaptive and coefficient of variance data fusion methods are used in this study. The pH electrode is fabricated with ruthenium dioxide (RuO2) sensing membrane using a sputtering system to deposit it onto a silicon substrate, and eight RuO2 pH electrodes are fabricated to form a sensor array for this study.

Liquid metal walls in fusion reactors will be subject to instabilities, turbulence, induced currents, error fields and temperature gradients that will make them locally bulge, thus entering in contact with the plasma, or deplete, hence exposing the underlying solid substrate. To prevent this, research has begun to actively stabilize static or flowing liquid metal layers by locally applying forces in feedback with thickness measurements. Here we present resistive sensors of liquid metal thickness and demonstrate jxB actuators, to locally control it.

In this paper, we address the problem of multiple view data fusion in the presence of noise and interferences. Recent studies have approached this problem using kernel methods, by relying particularly on a product of kernels constructed separately for each view. From a graph theory point of view, we analyze this fusion approach in a discrete setting. More specifically, based on a statistical model for the connectivity between data points, we propose an algorithm for the selection of the kernel bandwidth, a parameter, which, as we show, has important implications on the robustness of this fusion approach to interferences. Then, we consider the fusion of audio-visual speech signals measured by a single microphone and by a video camera pointed to the face of the speaker. Specifically, we address the task of voice activity detection, i.e., the detection of speech and non-speech segments, in the presence of structured interferences such as keyboard taps and office noise. We propose an algorithm for voice activity detection based on the audio-visual signal. Simulation results show that the proposed algorithm outperforms competing fusion and voice activity detection approaches. In addition, we demonstrate that a proper selection of the kernel bandwidth indeed leads to improved performance.

We consider a simple, yet widely studied, set-up in which a Fusion Center (FC) is asked to make a binary decision about a sequence of system states by relying on the possibly corrupted decisions provided by byzantine nodes, i.e. nodes which deliberately alter the result of the local decision to induce an error at the fusion center. When independent states are considered, the optimum fusion rule over a batch of observations has already been derived, however its complexity prevents its use in conjunction with large observation windows. In this paper, we propose a near-optimal algorithm based on message passing that greatly reduces the computational burden of the optimum fusion rule. In addition, the proposed algorithm retains very good performance also in the case of dependent system states. By first focusing on the case of small observation windows, we use numerical simulations to show that the proposed scheme introduces a negligible increase of the decision error probability compared to the optimum fusion rule. We then analyse the performance of the new scheme when the FC make its decision by relying on long observation windows. We do so by considering both the case of independent and Markovian system states and show that the obtained performance are superior to those obtained with prior suboptimal schemes. As an additional result, we confirm the previous finding that, in some cases, it is preferable for the byzantine nodes to minimise the mutual information between the sequence system states and the reports submitted to the FC, rather than always flipping the local decision.

This paper describes a distributed sensor-data-fusion paradigm and theory based on a previously developed theory to model sensors as highly autonomous units. Generic procedures are defined to reason and make decisions at the qualitative level. This facilitates distribution of intelligence ( code and hardware) to the sensor level and peer-to-peer communication among sensors, controllers, and other devices in the system.

A high-level data fusion system that adopts Bayesian statistics involving weights of evidence modelling is described to combine disparate information from airborne digital data such as a digital surface model (DSM), colour (ADS40), thermal infrared (TABI) and hyperspectral images (AISA) of different period. To show the efficacy of the system, an analysis of subtle change-detection is demonstrated. The data fusion system is capable of detecting changes in man-made features automatically in a densely populated area where there are no much prior information. Multi-class segmented images are obtained from the data captured by the four airborne remote sensing sensors. The system performs data fusion modelling by using binary images of each theme classes and thus a total of 40 binary patterns are obtained. Through Bayesian methods, involving weights of evidence modelling, all the binary images are analysed and finally four binary patterns (indicator images) are identified automatically as significant for the change-detection application. A weighted index overlay model available in the system, combines these four patterns. Data fusion by the weights of evidence modelling is found to be candid and unequivocal for predicting newly transformed locations. Results of Bayesian method are accurate as the weights are based on statistical analysis. Changes in features such as colour of roof, parking areas, openland areas, newly built structures, and presence or absence of vehicle is automatically extracted by using the high-level data fusion approach. The final predictor image showing the probability of change-detected areas in a high-dense city of Japan is generated robustly. The fused image products can be used to update maps for digital cartographic applications. ©2005 Microsoft Corporation. All rights reserved.

This article is from Sensors (Basel, Switzerland) , volume 10 . Abstract In this article we explain the architecture for the environment and sensors that has been built for the European project URUS (Ubiquitous Networking Robotics in Urban Sites), a project whose objective is to develop an adaptable network robot architecture for cooperation between network robots and human beings and/or the environment in urban areas. The project goal is to deploy a team of robots in an urban area to give a set of services to a user community. This paper addresses the sensor architecture devised for URUS and the type of robots and sensors used, including environment sensors and sensors onboard the robots. Furthermore, we also explain how sensor fusion takes place to achieve urban outdoor execution of robotic services. Finally some results of the project related to the sensor network are highlighted.