Method of Multi-Mode Sensor Data Fusion with an Adaptive Deep Coupling Convolutional Auto-Encoder

To address the difficulties in fusing multi-mode sensor data for complex industrial machinery, an adaptive deep coupling convolutional auto-encoder (ADCCAE) fusion method was proposed. First, the multi-mode features extracted synchronously by the CCAE were stacked and fed to the multi-channel convolution layers for fusion. Then, the fused data was passed to all connection layers for compression and fed to the Softmax module for classification. Finally, the coupling loss function coefficients and the network parameters were optimized through an adaptive approach using the gray wolf optimization (GWO) algorithm. Experimental comparisons showed that the proposed ADCCAE fusion model was superior to existing models for multi-mode data fusion.

Multisensor Data Fusion for High Quality Data Analysis and Processing in Measurement and Instrumentation

Multisensor data fusion （MDF） is an emerging technology to fuse data from multiple sensors in order to make a more accurate estimation of the environment through measurement and detection. Applications of MDF cross a wide spectrum in military and civilian areas. With the rapid evolution of computers and the proliferation of micro-mechanical/electrical systems sensors, the utilization of MDF is being popularized in research and applications. This paper focuses on application of MDF for high quality data analysis and processing in measurement and instrumentation. A practical, general data fusion scheme was established on the basis of feature extraction and merge of data from multiple sensors. This scheme integrates artificial neural networks for high performance pattern recognition. A number of successful applications in areas of NDI （Non-Destructive Inspection） corrosion detection, food quality and safety characterization, and precision agriculture are described and discussed in order to motivate new applications in these or other areas. This paper gives an overall picture of using the MDF method to increase the accuracy of data analysis and processing in measurement and instrumentation in different areas of applications.

Research on Kalman-filter based multisensor data fusion

Multisensor data fusion has played a significant role in diverse areas ranging from local robot guidance to global military theatre defense etc. Various multisensor data fusion methods have been extensively investigated by researchers, of which Klaman filtering is one of the most important. Kalman filtering is the best-known recursive least mean-square algorithm to optimally estimate the unknown states of a dynamic system, which has found widespread application in many areas. The scope of the work is restricted to investigate the various data fusion and track fusion techniques based on the Kalman Filter methods, then a new method of state fusion is proposed. Finally the simulation results demonstrate the effectiveness of the introduced method.

STUDY ON THE COAL-ROCK INTERFACE RECOGNITION METHOD BASED ON MULTI-SENSOR DATA FUSION TECHNIQUE

The coal-rock interface recognition method based on multi-sensor data fusiontechnique is put forward because of the localization of single type sensor recognition method. Themeasuring theory based on multi-sensor data fusion technique is analyzed, and hereby the testplatform of recognition system is manufactured. The advantage of data fusion with the fuzzy neuralnetwork (FNN) technique has been probed. The two-level FNN is constructed and data fusion is carriedout. The experiments show that in various conditions the method can always acquire a much higherrecognition rate than normal ones.

Deep Learning Based Data Fusion for Sensor Fault Diagnosis and Tolerance in Autonomous Vehicles

Environmental perception is one of the key technologies to realize autonomous vehicles.Autonomous vehicles are often equipped with multiple sensors to form a multi-source environmental perception system.Those sensors are very sensitive to light or background conditions,which will introduce a variety of global and local fault signals that bring great safety risks to autonomous driving system during long-term running.In this paper,a real-time data fusion network with fault diagnosis and fault tolerance mechanism is designed.By introducing prior features to realize the lightweight network,the features of the input data can be extracted in real time.A new sensor reliability evaluation method is proposed by calculating the global and local confidence of sensors.Through the temporal and spatial correlation between sensor data,the sensor redundancy is utilized to diagnose the local and global confidence level of sensor data in real time,eliminate the fault data,and ensure the accuracy and reliability of data fusion.Experiments show that the network achieves state-of-the-art results in speed and accuracy,and can accurately detect the location of the target when some sensors are out of focus or out of order.The fusion framework proposed in this paper is proved to be effective for intelligent vehicles in terms of real-time performance and reliability.

Reinforcement Learning Based Data Fusion Method for Multi-Sensors

In order to improve detection system robustness and reliability, multi-sensors fusion is used in modern air combat. In this paper, a data fusion method based on reinforcement learning is developed for multi-sensors. Initially, the cubic B-spline interpolation is used to solve time alignment problems of multisource data. Then, the reinforcement learning based data fusion(RLBDF) method is proposed to obtain the fusion results. With the case that the priori knowledge of target is obtained, the fusion accuracy reinforcement is realized by the error between fused value and actual value. Furthermore, the Fisher information is instead used as the reward if the priori knowledge is unable to be obtained. Simulations results verify that the developed method is feasible and effective for the multi-sensors data fusion in air combat.

Data fusion of target characteristic in multistatic passive radar

Radar cross section(RCS)is an important attribute of radar targets and has been widely used in automatic target recognition(ATR).In a passive radar,only the RCS multiplied by a coefficient is available due to the unknown transmitting parameters.For different transmitter-receiver(bistatic)pairs,the coefficients are different.Thus,the recovered RCS in different transmitter-receiver(bistatic)pairs cannot be fused for further use.In this paper,we propose a quantity named quasi-echo-power(QEP)as well as a method for eliminating differences of this quantity among different transmitter-receiver(bistatic)pairs.The QEP is defined as the target echo power after being compensated for distance and pattern propagation factor.The proposed method estimates the station difference coefficients(SDCs)of transmitter-receiver(bistatic)pairs relative to the reference transmitter-receiver(bistatic)pair first.Then,it compensates the QEP and gets the compensated QEP.The compensated QEP possesses a linear relationship with the target RCS.Statistical analyses on the simulated and real-life QEP data show that the proposed method can effectively estimate the SDC between different stations,and the compensated QEP from different receiving stations has the same distribution characteristics for the same target.

Research on Data Fusion of Adaptive Weighted Multi-Source Sensor

Data fusion can effectively process multi-sensor information to obtain more accurate and reliable results than a single sensor.The data of water quality in the environment comes from different sensors,thus the data must be fused.In our research,self-adaptive weighted data fusion method is used to respectively integrate the data from the PH value,temperature,oxygen dissolved and NH3 concentration of water quality environment.Based on the fusion,the Grubbs method is used to detect the abnormal data so as to provide data support for estimation,prediction and early warning of the water quality.

Weight Data Fusion Based on Mutual Support Applied in Large Diameter Measurement

The on-line diameter measurement of larger axis workpieces is hard to achieve high precision detection, because of the bad environment of locale, the problem to amend the measuring error by non-uniform temperature field, and the difficulty to collimate and locate by usual method. By improving the measurement accuracy of larger axis accessories, it is useful to raise axis and hole＇s industry produce level. Because of the influence of complex environment in locale and some influential factors which are hard excluded from the large diameter measurement with multi-rolling-wheels method, the measurement results may not support or even contradict each other. To the situation, this paper puts forward a mutual support deviation distinguish data fusion method, including mutual support deviation detection and weight data fusion. The mutual support deviation detection part can effectively remove or weaken the unexpected impact on the measurement results and the weight data fusion part can get more accurate estimate result to the detected data. So the method can further improve the reliability of measurement results and increase the accuracy of the measurement system. By using the weight data fusion based on the mutual support （DFMS） to the simulation and experiment data, both simulation results and experiment results show that the method can effectively distinguish the data influenced by unexpected impact and improve the stability and reliability of measurement results. The new provided mutual support deviation distinguish method can be used to single sensor measurement and multi-sensor measurement, and can be used as a reference in the data distinguish of other area. The DFMS is helpful to realize the diameter measurement expanded uncertainty in 5 ×10^-6D or even higher when the measured axis workpiece＇s diameter is 1-5 m （ 1 m ≤ D ≤5 m ）.

Multi-sensor measurement and data fusion technology for manufacturing process monitoring:a literature review

Due to the rapid development of precision manufacturing technology,much research has been conducted in the field of multisensor measurement and data fusion technology with a goal of enhancing monitoring capabilities in terms of measurement accuracy and information richness,thereby improving the efficiency and precision of manufacturing.In a multisensor system,each sensor independently measures certain parameters.Then,the system uses a relevant signalprocessing algorithm to combine all of the independent measurements into a comprehensive set of measurement results.The purpose of this paper is to describe multisensor measurement and data fusion technology and its applications in precision monitoring systems.The architecture of multisensor measurement systems is reviewed,and some implementations in manufacturing systems are presented.In addition to the multisensor measurement system,related data fusion methods and algorithms are summarized.Further perspectives on multisensor monitoring and data fusion technology are included at the end of this paper.

Multi-Sensor Data Fusion Technologies for Blanket Jamming Localization

The localization of the blanket jamming is studied and a new method of solving the localization ambiguity is proposed. Radars only can acquire angle information without range information when encountering the blanket jamming. Netted radars could get position information of the blanket jamming by make use of radars' relative position and the angle information, when there is one blanket jamming. In the presence of error, the localization method and the accuracy analysis of one blanket jamming are given. However, if there are more than one blanket jamming, and the two blanket jamming and two radars are coplanar, the localization of jamming could be error due to localization ambiguity. To solve this confusion, the Kalman filter model is established for all intersections, and through the initiation and association algorithm of multi-target, the false intersection can be eliminated. Simulations show that the presented method is valid.

Determination of Surface Precipitation Type Based on the Data Fusion Approach

Hazardous events related to atmospheric precipitation depend not only on the intensity of surface precipitation,but also on its type.Uncertainty related to determination of the precipitation type(PT)leads to financial losses in many areas of human activity,such as the power industry,agriculture,transportation,and many more.In this study,we use machine learning(ML)algorithms with the data fusion approach to more accurately determine surface PT.Based on surface synoptic observations,ERA5 reanalysis,and radar data,we distinguish between liquid,mixed,and solid precipitation types.The study domain considers the entire area of Poland and a period from 2015 to 2017.The purpose of this work is to address the question:“How can ML techniques applied in observational and NWP data help to improve the recognition of the surface PT?”Despite testing 33 parameters,it was found that a combination of the near-surface air temperature and the depth of the warm layer in the 0-1000 m above ground level(AGL)layer contains most of the signal needed to determine surface PT.The accrued probability of detection for liquid,solid,and mixed PTs according to the developed Random Forest model is 98.0%,98.8%,and 67.3%,respectively.The application of the ML technique and data fusion approach allows to significantly improve the robustness of PT prediction compared to commonly used baseline models and provides promising results for operational forecasters.

Lift fin stabilizers based on data fusion with wavelet denoising technology

Fin stabilizers with fin-lift feedback control can shield the mapping error of calculation between the fin angle and fin lift force,which is in the fin stabilizer with fin-angle feedback control.In practice,there are some technical difficulties in lift fin stabilizers,such as lift force detection and lift force sensor installation,so it cannot achieve the good antirolling performance.Therefore,a fin stabilizer system with fin-lift/fin-angle integrated control is brought forward.Data fusion based on wavelet denoising technology is employed in the system,which combines lift with fin angle local information from two sensors with different frequency ranges in order to eliminate redundant and contradictory information,and using complementary information to obtain the relative integrity of the lift force signal.The system model is established in this paper,and the fusion signal and the antirolling performance of this model are simulated respectively.The result shows that the control system can meet the antirolling need in different sea situations.

A multi-source data fusion modeling method for debris flow prevention engineering

The Digital Elevation Model(DEM)data of debris flow prevention engineering are the boundary of a debris flow prevention simulation,which provides accurate and reliable DEM data and is a key consideration in debris flow prevention simulations.Thus,this paper proposes a multi-source data fusion method.First,we constructed 3D models of debris flow prevention using virtual reality technology according to the relevant specifications.The 3D spatial data generated by 3D modeling were converted into DEM data for debris flow prevention engineering.Then,the accuracy and applicability of the DEM data were verified by the error analysis testing and fusion testing of the debris flow prevention simulation.Finally,we propose the Levels of Detail algorithm based on the quadtree structure to realize the visualization of a large-scale disaster prevention scene.The test results reveal that the data fusion method controlled the error rate of the DEM data of the debris flow prevention engineering within an allowable range and generated 3D volume data(obj format)to compensate for the deficiency of the DEM data whereby the 3D internal entity space is not expressed.Additionally,the levels of detailed method can dispatch the data of a large-scale debris flow hazard scene in real time to ensure a realistic 3D visualization.In summary,the proposed methods can be applied to the planning of debris flow prevention engineering and to the simulation of the debris flow prevention process.

THE EVALUATION, DATA FUSION AND APPLICATION OF FY2E SST IN TROPICAL CYCLONES OVER NORTHWEST PACIFIC OCEAN

The daily FY2 E Sea Surface Temperature(SST) data from China National Satellite Meteorological Center(NSMC) was evaluated and compared with the Optimum Interpolation Sea Surface Temperature(OISST) data from US National Oceanic and Atmospheric Administration(NOAA) over Northwest Pacific Ocean(NPO) in this study. The results show that the distribution of FY2 E SST is close to OISST in tropical region over NPO, especially in typhoon active season, but the value of FY2 E SST is a little lower than that of OISST in tropical ocean, with the absolute deviation 1℃ lower and the relative deviation about 6% lower. The correlation coefficient between monthly FY2 E SST and monthly OISST is as high as 0.7, which passes the t-test at a significance level of 0.01. Based on the evaluation result, the merged SST_(FY)over NPO is calculated using a weighting function. Besides, Tropical Cyclone Heat Potential(TCHP_(FY)) is calculated and combined with the simulated sea temperature profile. From three years operational tests in NSMC, the merged SST_(FY)and TCHP_(FY)are shown to be good indexes in monitoring and predicting the intensity of tropical cyclones(TCs) over NPO.

PALSAR-FBS L-HH Mode and Landsat-TM Data Fusion for Geological Mapping

Characterized by lithological diversity and rich mineral resources, Benshangul-Gumuz National Regional State located in Asosa Zones, Western Ethiopia has been investigated for geological mapping and morpho-structural lineaments extraction using PALSAR (Phased Array type L-band Synthetic Aperture Radar ) Fine Beam Single (FBS) L-HH polarization and Landsat-5 TM (Thematic Mapper ) datasets. These data were preprocessed to retrieve ground surface reflectance and backscatter coefficients. To overcome the geometry acquisition between the two sensors, they were geometrically and topographically rectified using ASTER-V2 DEM. Intensity-Hue-Saturation, directional filters and automatic lineaments extraction were applied on the datasets for lithological units’ discrimination and structural delimitation for potential mineral exploration. The obtained results showed good relationship among the topographic morphology, rock-substrate, structural variations properties, and drainage network. The spectral variations were easily associated with lithological units. Likewise, the morpho-structural information highlighted in the PALSAR image was visible without altering the radiometric integrity of the details in TM bands through the fusion process. Moreover, predominant lineaments directions trending NE-SW, NS, and NW-SE were identified. Results of this study highlighted the importance of the PALSAR FBS L-HH mode and TM data fusion to enhance geological features and lithological units for mineral exploration particularly in tropical zones.

A Study of Multi-sensor Data Fusion System Based on MAS for Nutrient Solution Measurement

For complementarity and redundancy of multi-sensor data fusion (MSDF) system,it is an effective approach for multiple components measurement.In order to measure nutrient solution on-line,a dynamic and complex system under greenhouse environment,sensors should have intelligent properties including self-calibration and self-compensation. Meanwhile,it is necessary for multiple sensors to cooperate and interact for enhancing reliability of multi-sensor system. Because of the properties of multi-agent system (MAS),it is an appropriate tool to study MSDF system.This paper proposed an architecture of MSDF system based on MAS for the multiple components measurement of nutrient solution.The sensor agent's structure and function modules are analyzed and described in detail,the formal definitions are given,too.The relations of the sensors are modeled to implement reliability diagnosis of the multi-sensor system,so that the reliability of nutrient control system is enhanced.This study offers an effective approach for the study of MSDF.

Multi-slots Wideband Cooperative Spectrum Sensing Based on Weighed Data Fusion in Cognitive Radio

In order to avoid the interference to the primary user(PU), in this paper Cognitive Radio (CR) periodically senses the presence of PU, and during one period, CR can sense all the sub-channels based on weighed data fusion and then use all the idle channels decided by the coordinator. The local sensing time of CR is divided into multi-slots in which CR can sense any sub-channel. Through reasonably allocating the sensing slots and users by mathematic optimization, the proposed algorithm can improve the total throughput of CR. The optimization problem of the proposed scheme which seeks to maximize the throughput subject to the constraint of the detected performance of each sub-channel is proposed in order to choose the optimum local sense time and the number of the cooperative CRs. The simulation results indicate that the proposed scheme can obtain higher throughput than the conventional single-channel sense, and there are the optimum local sense time and the number of cooperative CRs to make the throughput reach maximum.

Image Processing on Geological Data in Vector Format and Multi-Source Spatial Data Fusion

The geological data are constructed in vector format in geographical information system (GIS) while other data such as remote sensing images, geographical data and geochemical data are saved in raster ones. This paper converts the vector data into 8 bit images according to their importance to mineralization each by programming. We can communicate the geological meaning with the raster images by this method. The paper also fuses geographical data and geochemical data with the programmed strata data. The result shows that image fusion can express different intensities effectively and visualize the structure characters in 2 dimensions. Furthermore, it also can produce optimized information from multi-source data and express them more directly.

Unequal-interval data fusion algorithm for inertial/gravity matching integrated navigation system

Inertial/gravity matching integrated navigation system can effectively improve the longendurance navigation ability of underwater vehicles.Through the analysis of the matching process,the problem of unequal-interval in matching trajectory is addressed by an unequal-interval data fusion algorithm which is based on the unequal-interval characteristics analysis of the matching trajectory.Compared with previously available methods,the proposed algorithm improves the location precision.In conclusion,simulations of the integrated navigation system demonstrated the effectiveness and superiority of the proposed algorithm.