2019 22th International Conference on Information Fusion (FUSION)最新文献

{"title":"Machine Learning-Enabled Adaptation of Information Fusion Software Systems","authors":"Gerald Fry, Tameem Samawi, Kenny Lu, A. Pfeffer, Curt Wu, Steve Marotta, Michael Reposa, Stephen Chong","doi":"10.23919/fusion43075.2019.9011385","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011385","url":null,"abstract":"Real-time control systems must fuse information from multiple sensors to perform mission tasks in dynamic environments. The volatility of these environments can cause sensor degradation or failure, reducing the accuracy and reliability of the information fusion. The fusion system must automatically adapt to these environmental changes to minimize their effect while still completing its objectives. Further complicating the situation, many systems contain non-adaptive legacy software, so a generic adaptation framework must be able to adapt the system's software without any semantic knowledge. In this paper, we present an approach that uses semantic-agnostic program transformation to expand the range of behavior of an Unmanned Underwater Vehicle (UUV) information fusion system and optimize its behavior against mission objectives using machine-learning techniques. An analysis of our approach showed that adapting a UUV with our adaptation framework resulted in a 50% increase in its ability to search for an object under water and return safely while perturbed by battery failures.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"103 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131509724","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Hyperspherical Deterministic Sampling Based on Riemannian Geometry for Improved Nonlinear Bingham Filtering","authors":"Kailai Li, F. Pfaff, U. Hanebeck","doi":"10.23919/fusion43075.2019.9011390","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011390","url":null,"abstract":"We present a novel geometry-driven scheme for generating equally weighted deterministic samples of Bingham distributions in arbitrary dimensions. Unlike existing approaches, our method provides flexibility in the sampling size with samples satisfying requirements of the unscented transform while approximating higher-order moments of the Bingham distribution. This is done by first using Dirac mixture approximation as a sampling scheme on the tangent plane at the mode with respect to the Bingham density via gnomonic projection. Subsequently, the tangent sigma points are retracted backwards to the hypersphere, after which an on-manifold moment correction is performed via Riemannian optimization. The proposed approach is further applied to quaternion Bingham filtering for recursive orientation estimations. Evaluation results show that the geometry-adaptive sampling scheme gives better tracking accuracy and robustness for nonlinear orientation estimations.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"220 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128940829","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Pattern Classification in Heterogeneous Domains Based on Evidence Theory (Poster)","authors":"Zhun-ga Liu, Guanghui Qiu, G. Mercier, Q. Pan","doi":"10.23919/fusion43075.2019.9011174","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011174","url":null,"abstract":"In pattern classification, there exists a challenging problem when there are no training patterns. In some real applications, there may exist some labeled data in other related domain (called source domain), and such labeled data can be helpful to solve the classification problem in target domain. It is considered that the source domain and target domain are heterogeneous here. A new heterogeneous data transfer classification method based on evidence theory is proposed. Some corresponding patterns (pattern pairs) in source domain and target domain are given to build the link of these two domains. For each pattern in target domain, it is hard to determine one exact mapping value in source domain due to the distinct characteristics of these two domains. So we estimate a mapping scope in source domain using KNN technique. The target pattern is allowed to have multiple mapping values in the scope with different weights/reliabilties. These mapping values can produce different classification results. Evidence theory is good at combining the uncertain information. Therefore a new weighted DS fusion method is developed for combining these classification results, which are discounted by the corresponding weights, and the final class decision is made according to the combination result. A pair of heterogeneous remote sensing images and some UCI data sets are used in this paper to test the performance of our method with respect to several other methods, and it shows that the new method can efficiently improve the classification accuracy.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"132 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130841733","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Maximum Likelihood Detection in Focal Plane Arrays with Generic Point Spread Function (Poster)","authors":"K. Kiani, B. Balasingam, B. Shahrrava","doi":"10.23919/fusion43075.2019.9011223","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011223","url":null,"abstract":"In this paper the problem of target detection on images and focal plane arrays (FPA) is considered. The proposed approach has applications in biomedical systems, autonomous surveillance systems, target tracking systems and robotics. In a previous paper at the Fusion conference the problem of single, point target detection on FPA was solved under the assumption that the point spread function of the target is strictly circular. The proposed approach in this paper extends the previous result for a generic point spread function that is more applicable to solve practical problems. In this paper, we derive the maximum likelihood (ML) target detector for image observations; the proposed ML detector is optimal under the generic assumption that the FPA contains a single target that is in the form of a Gaussian signal intensity with known covariance Matrix. Further, we derive the Cramer-Rao lower bound (CRLB) of the estimation and then present the hypothesis test to find a threshold for target acceptance. Finally, we theoretically derive the receiver operating characteristic (ROC) curve of the detector. Simulation results show that the ML estimator is efficient and that the theoretically derived ROC is a close approximation to the realistic one at very low signal to noise ratio values.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"167 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133532803","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Coursa Venue: Indoor Navigation Platform Using Fusion of Inertial Sensors with Magnetic and Radio Fingerprinting","authors":"Gennadii Berkovich, Dmitry Churikov, J. Georgy, C. Goodall","doi":"10.23919/fusion43075.2019.9011323","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011323","url":null,"abstract":"Navigation is one of the most essential human needs. People need to know their location anywhere and everywhere, indoors and outdoors, for example for tracking children and older people who are in extreme need of care or for providing a precise indoor location of calls to 911. The latter is especially important since more than seventy percent of calls are generated indoors. Other important fields are tracking of personnel in possibly dangerous environment, LBS/LBA applications, and social networking that rely on indoor localization. Unfortunately, broadly used satellite navigation receivers work perfectly only under open sky. Operating indoors, GPS/GNSS receivers suffer from signal attenuation when satellite signals propagate through a roof and walls of a building and from the multipath due to their reflection. As a result, accurate GNSS-based position fix is almost impossible in most indoor conditions. We present Coursa Venue solution that was developed by TDK-Invensense for infrastructure-less indoor positioning on commercial smartphones. The solution consists of two major parts: cloud-based software for fingerprinting and mobile applications for Android and iOS that provide real-time blue dot positions to users. TDK-Invensense's approach to real-time indoor positioning is based on fusion of multiple technologies. Measurements from such smartphone sensors as IMU (3D accelerometer, gyroscope), a magnetic field sensor (3D magnetometer), WiFi and BLE modules are used for hybrid indoor positioning in the navigation engine. Particle filtering is used as the fusion engine. Indoor navigation software uses such technologies as PDR, geomagnetic fingerprinting, Wi-Fi/BLE fingerprinting, and, optionally, map matching. TDK-Invensense' ‘s PDR provides prediction of user relative movement regardless of orientation and misalignment of a smartphone, whereas magnetic and radio fingerprinting serves for correction of inertial sensors error. The cloud-based component can create magnetic and radio fingerprint databases using either data collected by designated surveyors who walked inside a venue by predetermined routes, or crowdsourced data from users of a real-time mobile application collected during their everyday activity. This paper discusses the architecture of the Coursa Venue solution and demonstrates its positioning results in several venues with comparison to ground truth paths to provide statistical assessment and key performance indices.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"8 1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115676078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-Sensor Passive Localization Based on Sensor Selection","authors":"Wen Ma, Hongyan Zhu, Yan Lin","doi":"10.23919/fusion43075.2019.9011312","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011312","url":null,"abstract":"In passive localization applications, the positioning accuracy for an emitter is highly dependent on the geometry between sensors and the target, the site error and measurement noise of sensors. We propose a sensor selection mechanism which aims to choose a subset of sensors to implement the multi-sensor passive localization. An optimization model is established by minimizing the GDOP (geometric dilution of precision), with or without the constraint on the cardinality of the selected sensor subset. The CEO (cross entropy optimization) is employed to solve the resulting complex combinatorial optimization model. Simulation experiments are conducted and simulation results demonstrate the efficiency of the proposed sensor selection scheme for multi-sensor passive localization.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124148969","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mixed Open-and-Closed Loop Satellite Task Planning","authors":"R. Tharmarasa, T. Kirubarajan, J. Berger, M. Florea","doi":"10.23919/fusion43075.2019.9011405","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011405","url":null,"abstract":"A key problem in earth observation using satellite-based sensors is to allocate the limited resources to achieve optimal performance. In the literature, several algorithms are proposed to handle this problem as an open-loop problem. Using a closed-loop formulation over an open-loop formulation would provide a better solution if the closed-loop solution could be implemented in real-time. However, an optimal closed-loop formulation may not be able to handle a medium or a large scale problem in real-time. In this work, we have proposed a mixed open-and-closed loop algorithm to handle the problem in real-time by taking the advantages of open- and closed-loop formulations. Simulation results illustrating the performance of the proposed algorithm are also presented.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"170 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114359924","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Inertial Motion Capture Using Adaptive Sensor Fusion and Joint Angle Drift Correction","authors":"H. T. Butt, Manthan Pancholi, Mathias Musahl, Pramod Murthy, M. A. Sanchez, D. Stricker","doi":"10.23919/fusion43075.2019.9011359","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011359","url":null,"abstract":"The ambulatory motion capture and gait analysis using wearable MEMS based magnetic-inertial measurement units (MIMUs) is challenging despite multisensor fusion and effective anatomical (sensor-to-segment) calibration. The MEMS based sensors show degraded performance when run for long time, especially indoors. This is due to the fact that assumption of no acceleration except gravity and homogenous magnetic field no longer holds, when the motion is being performed. The rate gyro is used to complement the accelerometer/ magnetometer for orientation estimation, but integration of its residual biases as well as noise eventually causes the sensor fusion estimates to drift. The errors in heading angle or yaw are particular significant due to persistent nature of magnetic inhomogeneity in the environment. This ultimately results in inaccurate and drifting joint angle estimates between body segments that would require some means of correction. In present work, we propose a new adaptive covariance based EKF for sensor fusion which makes it effectively robust to both dynamic body accelerations as well as inhomogeneous magnetic field. The adaptive covariance method penalizes the bad accelerometer and magnetometer measurements and intelligently updates the gyro biases online using only undisturbed readings of accelerometer/magnetometer. Our sensor fusion algorithm provides accurate orientation estimates for each MIMU node over time. In order to account for any residual drift of joint angles, we propose a novel correction term in our anatomical formulation that performs online correction of drift in individual joint angles and updates it as an orientation offset. This offset correction for joint angle is performed automatically when the limb or extended torso are in neutral quasi-static pose and this condition is judged using accelerometers. Overall our approach achieves precise orientation estimates in highly dynamic conditions and avoids drift or error accumulation due to inhomogeneous magnetic fields during inertial motion capture.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114946382","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated Local Linearization Particle Filter for Multiple Maneuvering Target Tracking in Clutter","authors":"Seung-Hyo Park, T. Song, S. Chong","doi":"10.23919/fusion43075.2019.9011256","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011256","url":null,"abstract":"The integrated particle filter (IPF) algorithm is proposed for single target tracking in clutter that combines the existing particle filters with false track discrimination (FTD) which distinguishes between the true tracks and the false tracks using the target existence probability as a track quality measure. To improve the tracking performance of IPF for maneuvering multitarget tracking, we propose an integrated local linearization particle filter (ILLPF) that applies the FTD to LLPF which approximates the optimal importance density with the updated estimates of a bank of tracking filters. The proposed algorithm is extended to accommodate interacting multiple model-linear multitarget-ILLPF (IMM-LM-ILLPF) for maneuvering target tracking with multiple target dynamic models for robust tracking. A study with Monte Carlo simulation demonstrates the improvement of maneuvering multitarget tracking performance in cluttered environments.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"32 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128461238","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Object Detection in Tensor Decomposition Based Multi Target Tracking","authors":"F. Govaers","doi":"10.23919/fusion43075.2019.9011185","DOIUrl":"https://doi.org/10.23919/fusion43075.2019.9011185","url":null,"abstract":"Non-linear filtering arises in many sensor applications such as for instance robotics, military reconnaissance, advanced driver assistance systems and other safety and security data processing algorithms. Since a closed-form of the Bayesian estimation approach is intractable in general, approximative methods have to be applied. Kalman or particle based approaches have the drawback of either a Gaussian approximation or a curse of dimensionality which both leads to a reduction in the performance in challenging scenarios. An approach to overcome this situation is state estimation using decomposed tensors. In this paper the Sequential Likelihood Ratio Test (SLRT) for object detection in tensor decomposition based target tracking is presented. The scheme closely follows the well-known and often applied approach of the track-oriented MHT.","PeriodicalId":348881,"journal":{"name":"2019 22th International Conference on Information Fusion (FUSION)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129617070","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}