Search

Video-based navigation is an increasingly used procedure with hard real-time requirements and high computational effort. In this field, FPGA hardware acceleration supplies low-cost and considerable performances enhancement. Video-based navigation algorithms extrapolate and correlate features from images, relying on their accuracy. Image enhancement provides more defined and contrasted frames, assuring high precision feature extraction. The paper introduces an FPGA-based self-adaptive image enhancer. The IP-core is suitable for hard-real time applications, such as space applications, thanks to the guaranteed high-throughput.

International audience; The system of heart rate/cycling power can be modeled through a nonlinear Hammerstein ARX model with uncertainties. Therefore, we investigate these models via the so-called Takagi-Sugeno (TS) framework. TS models belong to the family of quasi-LPV and allow an exact representation of the nonlinear system. The goal of the paper is to show that an efficient output feedback control can be achieved via specific Lyapunov functions. Results from real-time application show the interest of the proposed approach.

This chapter will explore the potential role that naturopaths may play in public health care, where appropriate, and discuss relevant practice, regulatory, and legislative developments that should be part of any naturopathic role in public health care delivery

Bu çalışma, 07-08 Haziran 2010 tarihleri arasında Bursa[Türkiye]’da düzenlenen 5. Automotive Technology Conference (OTEKON)’da bildiri olarak sunulmuştur. The objective of this study is to design storage assignment and order picking system using a developed mathematical model and stochastic evolutionary optimization approach in the automotive industry. It is performed in two stages. At the first stage, storage location assignment problem is solved with a class-based storage policy with the aim of minimizing warehouse transmissions by using integer programming. At the second stage, batching and routing problems are considered together to minimize travel cost in warehouse operations. A warehouse in the automotive industry is analyzed, and an optimum solution is obtained from an integer programming model. Due to the computational time required for solving the integer programming problem, a faster genetic algorithm is also developed to form optimal batches and optimal routes for the order picker. The main advantage of the algorithm is the quick response to production orders in real-time applications. The solutions showed that the proposed approach based on genetic algorithms can be applied and integrated to any kind of warehouse layout in automotive industry.

Achieving effective facial emotional expressivity within a real-time rendering constraint requests to leverage on all possible inspiration sources and especially from the observations of real individuals. One of them is the frequent asymmetry of facial expressions of emotions, which allows to express complex emotional feelings such as suspicion, smirk, and hidden emotion due to social conventions. To achieve such a higher degree of facial expression, we propose a new model for mapping emotions onto a small set of 1D Facial Part Actions (FPA)s that act on antagonist muscle groups or on individual head orientation degree of freedoms. The proposed linear model can automatically drive a large number of autonomous virtual humans or support the interactive design of complex facial expressions over time.

Due to the recursive and integrative nature of zero-velocity-update-aided (ZUPT-aided) inertial navigation systems (INSs), the error covariance increases throughout each ZUPT-less period followed by a drastic decrease and large state estimate corrections as soon as ZUPTs are applied. For dead-reckoning with foot-mounted inertial sensors, this gives undesirable discontinuities in the estimated trajectory at the end of each step. However, for many applications, some degree of lag can be tolerated and the information provided by the ZUPTs at the end of a step can be made available throughout the step, eliminating the discontinuities. For this purpose, we propose a smoothing algorithm for ZUPT-aided INSs. For near real-time applications, smoothing is applied to the data in a step-wise manner requiring a suggested varying-lag segmentation rule. For complete off-line processing, full data set smoothing is examined. Finally, the consequences and impact of smoothing are analyzed and quantified based on real-data. QC 20130114

Co-encadrement de la thèse : Bruno Steux; This work is part of the AROS project. Its goal is to define a fast prototyping tool for dynamic and distributed real-time applications, mostly for automotive industry and robotic. Two distinct methods are normally used to develop distributed real-time applications. The first one -the time triggered approach- is based on worst execution time analysis, whereby time sharing for the various tasks of an application is statically defined. This approach offers considerable safety but the time analysis is sometimes difficult to process. The second one -the priority scheduling approach- is based on ascribing a priority level to each task, which will then allow the system to define an execution order, based on the events is has received. This second approach is more flexible and easier to implement but is less safe and cannot ensure that the application behaves predictably. The structure of the AROS applications being dynamic, the time-triggered approach is irrelevant as it requires a static analysis that cannot be conducted. The priority scheduling approach is also irrelevant because of the non predictable behaviour. We propose an approach based on causal events scheduling inspired by distributed event simulators scheduling techniques. While comparatively easy to use for application designers, this new approach produces applications with a perfectly predictable behaviour. Two main obstacles must be overcome: the real time synchronisation of the execution engine and compliance with real-time constraints.; Ces travaux s'inscrivent dans le cadre du projet AROS (Automotive Robust Operating Services). Il a pour objectif de proposer un outil de prototypage rapide d'applications dynamiques distribuées temps-réel, principalement dans le domaine de l'automobile et de la robotique. Les applications distribuées temps-réel sont traditionnellement développées selon deux approches. La première, l'ordonnancement temporel, est basée sur l'analyse du pire temps d'exécution (worst execution time). Un partage du temps entre les différentes tâches de l'application est établi de façon statique. Cette technique offre une grande sureté de fonctionnement au prix d'une analyse temporelle parfois difficile à mener. La seconde, l'ordonnancement par priorité, est basée sur l'attribution à chaque tâche d'un niveau de priorité qui permet d'établir l'ordre d'exécution en fonction des évènements reçus par le système. Cette seconde technique, plus souple à mettre en œuvre, offre moins de garanties et conduit à un comportement non déterministe de l'application. La structure des applications AROS étant dynamique, l'approche temporelle est exclue car elle demande une analyse statique qu'il est impossible de produire. L'approche basée sur les priorités d'exécution est également exclue à cause de son non déterminisme comportemental. Nous proposons une approche basée sur un ordonnancement évènementiel causal inspirée des techniques d'ordonnancement des simulateurs évènementiels distribués. Tout en étant relativement simple à utiliser pour le concepteur d'application, cette technique produit des applications dont le comportement est parfaitement déterministe. Deux principales difficultés sont à surmonter : la synchronisation en temps-réel du moteur d'exécution et le respect des contraintes temps-réel.

In this paper we present some preliminary results on the application of Compressive Sensing (CS) to high resolution radar imaging. CS is a recently developed theory which allows reconstruction of sparse signals with a number of measurements much lower than what is required by the Shannon sampling theorem. This method has already found its way in a number of applications where the sampling rate or the acquisition time are prohibitive for real time applications, such as high resolution medical and optical imaging. Actual demonstrations of CS with experimental radar data are still very few, and therefore in this paper we apply CS to two-dimensional radar imaging with experimental data. The measurement setup contains a small number of corner reflectors which are illuminated using a stepped sequence of frequencies, over a range of aspect angles. The CS approach uses only a random selection of frequencies and angles, to reconstruct the two-dimensional image. The results obtained with CS are compared with the one achieved with conventional focusing (Matched Filter). The results show that the corner reflectors are properly reconstructed, with a significant reduction in the amount of measurement samples.

Conference name: First International Workshop, HBU 2010 Date of Conference: August 22, 2010 This paper provides a comparative study on the different techniques of classifying human activities that are performed using body-worn miniature inertial and magnetic sensors. The classification techniques implemented and compared in this study are: Bayesian decision making (BDM), the least-squares method (LSM), the k-nearest neighbor algorithm (k-NN), dynamic time warping (DTW), support vector machines (SVM), and artificial neural networks (ANN). Daily and sports activities are classified using five sensor units worn by eight subjects on the chest, the arms, and the legs. Each sensor unit comprises a triaxial gyroscope, a triaxial accelerometer, and a triaxial magnetometer. Principal component analysis (PCA) and sequential forward feature selection (SFFS) methods are employed for feature reduction. For a small number of features, SFFS demonstrates better performance and should be preferable especially in real-time applications. The classifiers are validated using different cross-validation techniques. Among the different classifiers we have considered, BDM results in the highest correct classification rate with relatively small computational cost. © 2010 Springer-Verlag Berlin Heidelberg.

Date of Conference: 22-24 April 2010 As computer generated holograms becomes more common, the fast computation of holographic interference patterns in digital environment becomes a necessity. Since the computation time of holograms via Fresnel (or Rayleigh-Sommerfeld) diffraction models makes real time applications impossible, the holographic stereograms are developed to be a solution for this problem. Holographic stereograms divide the hologram plane into segments. In phase added stereograms the coordinates of 3D source points are used while calculating the diffraction field. And that enables to calculate the diffraction field with appropriate sized FFTs. Although the phase added stereograms are advantageous in terms of computation time, the quality of the reconstructed three dimensional images may not be satisfactory. The main reason is that the diffraction field of a given point source is approximated as a pure complex sinusuoid in each segment. To increase the reconstruction quality, we propose a method that uses variable sized segments, as opposed to previously developed holographic stereograms that use fixed sized segments. While approximating the diffraction field of a point source, higher frequency regions are covered with smaller segments and lower frequency regions with larger segments. As a result of this, we keep the total number of oscillations of pure sinusoidal waves constant in each segment. The simulations that we carried out for a point source show that we are able to obtain better quality reconstruction with our method. ©2010 IEEE.

International audience; In this paper, we present a generic framework for urban vehicle navigation using a topological map. This map is built by taking into account the non-holonomic behaviour of the vehicle. After a localization step, a sensory route is extracted to reach a goal. This route is followed using a sensor-based control strategy, based on the vehicle model and computed from the state extracted from the current and the desired sensory images. In that aim, a generic model is proposed for visual sensors. Experiments with an urban electric vehicle navigating in an outdoor environment have been carried out with a fisheye camera using a single camera and natural landmarks. A navigation along a 1700-meter-long trajectory validates our approach.

Power system security poses an important issue in planning and operation stages of a power system. Conventional method of security evaluation involves solving full AC load flow equations and transient stability analysis by time domain simulation. This technique is highly time consuming and infeasible for real time applications. The Pattern Recognition (PR) approach is recognized as an alternative tool for online security evaluation. This paper proposes a Support Vector Machine (SVM) based classification for transient security assessment. The proposed SVM-PR approach is tested on 9 Bus WSCC, 30 Bus and 57 Bus IEEE standard systems and the transient security status of the system is accessed. The performance of the SVM classifier is gauged in terms of measures like classification accuracy, execution time and misclassification rate and the results are compared with the Multilayer Perceptron (MLP) classifier. The proposed method helps to identify the most dominant system dependent features responsible for transient security classification and yields fairly high classification accuracy with less computation time. Further, the proposed approach can realistically be applied to large scale systems, as the number of input attributes selected for classification is independent of the system size. Copyright � JES 2009.

International audience; This paper presents a modified-octree technique that generates a mixedelement mesh. The final output mesh consider cubes, prisms, pyramids and tetrahedra. This technique is optimized for brain tumor extraction simulation in a real-time application. The proposed technique is based on the octree algorithm with a specific constraint: elements will be split only if they intersects a certain region of interest. With this approach we pursued a refined mesh only in the path from the skull opening point to the tumor. Fast computation by the Finite Element Method (FEM) is achieve thanks to the local refinement. Examples are given and comparison with other approaches are presented.

The flexibility of current graphics hardware is still not enough to ensure the full implementation of an original complex algorithm such as a local tone mapping operator, which maintains the same quality performances as its original CPU implementation. Significant changes are often needed to the original CPU implementation in order to overcome many of the limitations of the current graphics hardware. As a result of this we often have reduced quality reproduction, and the frame rate of the GPU implementation is not always acceptable for real-time applications. In this paper, we show how to change the CPU implementation of a state of the art local tone mapping operator for accelerating the computation process to real time frame rates. We also present a modification of the luminance local adaptation computation, showing a simple but not yet exploited property of the Gaussian filter, allowing us to maintain the same quality appearance of the original tone mapping operator. Finally we test the hardware implementation on NVIDIA graphics cards on several images and as well as a video. We compare our hardware implementation with the corresponding CPU implementation and previous work. Copyright © 2007 by the Association for Computing Machinery, Inc. 213 218 Sponsors: Comenius University Conference code: 114606 Cited By :6

Quality of Service (QoS) support in the current internet is indispensable because of QoS-sensitive real-time applications such as Voice-over-IP, IP-TV, video conferencing, online gaming etc. Since the introduction of the Differentiated Services (Diff-Serv) architecture there has been considerable work reported in literature on its performance evaluation. However, none of them have addressed the basic issue of quantification of QoS for supporting streaming media. The main contribution in this paper is the quantification of the QoS of streaming media in terms of Mean Opinion Score (MOS) values. A test bed has been implemented using off-the-shelf components. The experimental results and MOS values are used to show that in a DiffServ Assured Forwarding network architecture, with class based weighted fair queue scheduling discipline, the QoS of streaming media is not compromised when the load exceeds the reserved capacity, even in case of congestion.

Cam, Ertugrul/0000-0001-6491-9225 WOS: 000242103800020 In this paper, an adaptive controller was applied to a 765 MW large thermal power plant to decrease operating costs, increase quality of generated electricity and satisfy environmental concerns. Since power plants may present several operating problems such as disturbances and severe effects at operating points, design of their controllers needs to be carried out adequately. For these reasons, first, a reduced mathematical model was developed under Computer Aided Analysis and Design Package for Control, (CADACS), so that the results of the experimental model have briefly been discussed. Second, conventional PID and adaptive controllers were designed and implemented under the real-time environment of the CADACS software. Additionally, the design of the adaptive model-reference and conventional PID controllers used in the power plant for real-time control were theoretically presented. All processes were realized in real-time. Due to safety restrictions, a direct connection to the sensors and actuators of the plant was not allowed. Insted a coupling to the control system was realized. This offers, in addition, the usage of the supervisiory functions of an industrial process computer system. Application of the controllers indicated that the proposed adaptive controller has better performances for rise and settling times of electrical power, and enthalpy outputs than the conventional PID controller does. (c) 2006 Elsevier Ltd. All rights reserved.

Quality of service (QoS) support in the current Internet is indispensable because of QoS-sensitive real-time applications such as voice-over-IP, IP-TV, video conferencing, online gaining etc. Since the introduction of the differentiated services (Diff-Serv) architecture there has been considerable work reported in literature on its performance evaluation. However, none of them have addressed the basic issue of quantification of QoS for supporting streaming media. The main contribution in this paper is the quantification of the QoS of streaming media in terms of mean opinion score (MOS) values. A test bed has been implemented using off-the-shelf components. The experimental results and MOS values are used to show that in a DiffServ assured forwarding network architecture, with class based weighted fair queue scheduling discipline, the QoS of streaming media is not compromised when the load exceeds the reserved capacity, even in case of congestion.

In classical gasoline injection systems, the control of the throttle plate position is important for efficiency and emission of the engine. The angular position, depending on the current load of the engine, must track a desired trajectory which is performed by the accelerator. In Feedback linearization approach the state space must be available: measurable or observable. If the system is nonlinear, observers of the state in real time application are often difficult to be implemented. A technique to approximate derivatives is proposed in order to avoid velocity and acceleration measurements. The tracking problem is treated by using Kharitonov's approach. Thanks to an approximated PD regulator robustness in the proposed feedback linearization is achieved. Measured results on the real experimental set point are reported. In classical gasoline injection systems, the control of the throttle plate position is important for efficiency and emission of the engine. The angular position, depending on the current load of the engine, must track a desired trajectory which is performed by the accelerator. In Feedback linearization approach the state space must be available: measurable or observable. If the system is nonlinear, observers of the state in real time application are often difficult to be implemented. A technique to approximate derivatives is proposed in order to avoid velocity and acceleration measurements. The tracking problem is treated by using Kharitonov's approach. Thanks to an approximated PD regulator robustness in the proposed feedback linearization is achieved. Measured results on the real experimental set point are reported.

The coding efficiency of the H.264/AVC standard enables the transmission of high quality video over bandwidth limited networks. Due to the use of multiple Macroblock (MB) partitions, the Motion estimation module has extremely high complexity that makes it unpractical for most real-time applications on resource-limited platforms such as hand held devices. In this paper we propose a novel algorithm that significantly reduces the encoding complexity while maintaining high rate distortion performance. The proposed method reduces the Motion estimation (ME) computational complexity by accurately predicting the optimal MB partitions and restricting the number of candidate modes based on a-priori probabilities computed from spatio-temporal information. The experimental results show that the speed up of UmHexagonS [1] (one of the most efficient ME algorithms) can be doubled while maintaining the coding efficiency of Full Search. Publisher's Version WOS:000245768500338

Video surveillance in outdoor areas is hampered by consistent background motion which defeats systems that use motion to identify intruders. While algorithms exist for masking out regions with motion, a better approach is to develop a statistical model of the typical dynamic video appearance. This allows the detection of potential intruders even in front of trees and grass waving in the wind, waves across a lake, or cars moving past. In this paper we present a general framework for the identification of anomalies in video, and a comparison of statistical models that characterize the local video dynamics at each pixel neighborhood. A real-time implementation of these algorithms runs on an 800 MHz laptop, and we present qualitative results in many application domains.

During the last years the department of Optical Information Systems of the German Aerospace Center (DLR) developed a considerable number of imaging sensor systems for a wide field of applications. Systems with a high geometric and radiometric resolution in dedicated spectral ranges of the electromagnetic spectrum were provided by developing and applying cutting edge technologies. Designed for photogrammetry and remote sensing, such systems play an important role for security and defence tasks. Complete system solutions were implemented considering theoretical framework, hardware design and deployment, overall system tests, calibration, sensor operation and data processing. Outstanding results were achieved with the airborne digital sensor ADS40 and the micro satellite BIRD and its infrared camera payload. Future activities will focus on intelligent cameras and sensor webs. The huge amount of data will force the issue of thematic multi-sensor data processing which is to be implemented in real time near the sensor. In dependence on well defined tasks, combinations of several sensors with special properties will be placed on spaceborne, airborne or terrestrial platforms. The paper gives an overview about finished and current projects and strategic goals.

The paper presents a computational model for real-time control tasks, with the primary goal of simplifying the control and scheduling codesign problem. The model combines time-triggered I/O and inter-task communication with dynamic, reservation-based task scheduling. To facilitate short input-output latencies, a task may be divided into several segments. Jitter is reduced by allowing communication only at the beginning and at the end of a segment. A key property of the model is that both schedulability and control performance of a control task will depend on the reserved utilization factor only. This enables controllers to be treated as scalable real-time components. The model has been implemented in a real-time kernel and validated in a real-time control application.

presupposto indispensabile per l'analisi, la progettazione e la gestione di una qualsiasi tipologia di intervento sulla rete stradale è costituzione di una adeguata catena informativa sullo stato di funzionamento del sistema stesso, a cui generalmente si fa riferimento con il nome di "monitoraggio del traffico".

Colloque sur invitation. nationale.; National audience; In-vehicle embedded computer control systems are special cases of a distributed real-time system. In the design step these systems should be validated and optimised. As the part of the Franco-Chinese program PRA SI99-04, we presented during a seminar in Wuhan University of Technology (China) our research results on the validation techniques allowing us to verify the real-time constraints meeting of such systems. These techniques include modelling and performance evaluation based on modular simulation model building process and tool, schedulability analysis of periodic tasks and messages, combined stochastic and deterministic analysis of the system's robustness faced to transmission errors and dual-priority dynamic scheduling of periodic and non periodic messages. They are applied first to a single CAN networked system and then to a multiple networked system (CAN and VAN). The PRA SI99-04 project aiming to integrate security control (car colliding and fire extinction) to the existing embedded system in a car is outlined. Other related research activities developed in PREDIT-CAROSSE project (LORIA-PSA-LIP6) and in AEE ("Architecture Electronique Embarquée") national project are brief described.

An active distribution network (ADN) is an electrical-power distribution network that implements a real-time monitoring and control of the electrical resources and the grid. Effective monitoring and control is realised by deploying a large number of sensing and actuating devices and a communication network to facilitates the two-way transfer of data. The reliance of ADN operations on a large number of electronic devices and on communication networks poses a challenge in protecting the system against cyber-attacks. Identifying these challenges and commissioning appropriate solutions is of utmost importance to realize the full potential of a smart grid that seamlessly integrates distributed generation, such as renewable energy sources. As a first step, we perform a thorough threat analysis of a typical ADN. We identify potential threats against field devices, the communication infrastructure and servers at control centers. We also propose a check-list of security solutions and best practices that guarantee a distribution network's resilient operation in the presence of malicious attackers, natural disasters, and other unintended failures that could potentially lead to islanded communication zone. For the next step, we investigate the security of MPLS-TP, a technology that is mainly used for long-distance inter-domain communication in smart grid. We find that an MPLS-TP implementation in Cisco IOS has serious security vulnerabilities in two of its protocols, BFD and PSC. These two protocols control protection-switching features in MPLS-TP. In our test-bed, we demonstrate that an attacker who has physical access to the network can exploit the vulnerabilities in order to inject forged BFD or PSC messages that affect the network's availability. Third, we consider multicast source authentication for synchrophasor data communication in grid monitoring systems (GMS). Ensuring source authentication without violating the stringent real-time requirement of GMS is challenging. Through an extensive review of existing schemes, we identified a set of schemes that satisfy some desirable requirements for GMS. The identified schemes are ECDSA, TV-HORS and Incomplete- key-set. We experimentally compared these schemes using computation, communication and key management overheads as performance metrics. A tweak in ECDSA's implementation to make it use pre-generated tokens to generate signatures significantly improves the computation overhead of ECDSA, making it the preferred scheme for GMS. This finding is contrary to the generally accepted view that asymmetric cryptography is inapplicable for real-time systems. Finally, we studied a planning problem that arises when a utility wants to roll out a software patch that requires rebooting to all PMUs while maintaining system observability. The problem we address is how to find a partitioning of the set of the deployed PMUs into as few subsets as possible such that all the PMUs in one subset can be patched in one round while all the PMUs in the other subsets provide full observability. We show that the problem is NP-complete in the general case and and formulated it as binary integer linear programming (BILP) problem. We have also provided an heuristic algorithm to find an approximate solution. Furthermore, we have identified a special case of the problem where the grid is a tree and provided a polynomial-time algorithm that finds an optimal patching plan that requires only two rounds to patch the PMUs.

In this paper we present some preliminary results on the application of Compressive Sensing (CS) to high resolution radar imaging. CS is a recently developed theory which allows reconstruction of sparse signals with a number of measurements much lower than what is required by the Shannon sampling theorem. This method has already found its way in a number of applications where the sampling rate or the acquisition time are prohibitive for real time applications, such as high resolution medical and optical imaging. Actual demonstrations of CS with experimental radar data are still very few, and therefore in this paper we apply CS to two-dimensional radar imaging with experimental data. The measurement setup contains a small number of corner reflectors which are illuminated using a stepped sequence of frequencies, over a range of aspect angles. The CS approach uses only a random selection of frequencies and angles, to reconstruct the two-dimensional image. The results obtained with CS are compared with the one achieved with conventional focusing (Matched Filter). The results show that the corner reflectors are properly reconstructed, with a significant reduction in the amount of measurement samples.

cited By (since 1996) 0; Conference of 16th AIAA/DLR/DGLR International Space Planes and Hypersonic Systems and Technologies Conference; Conference Date: 19 October 2009 through 22 October 2009; Conference Code: 81964; International audience; In the framework of the hypersonic propulsion, with cooled engine by endothermic hydrocarbon fuel, it is necessary to provide adapted measurement methods for the cooling regulation as for the control of the engine thrust. The sensors should be robust because of the extreme in-flight conditions, of vehicle acceleration and vibration, of the fluid temperature (1500 K) and pressure (3.5 MPa) and of the multi-component supercritical mixture. Their response time should be lower than one second. Therefore, a large range of real time and on line measurement methods has been tested and some of these methods among the most promising are presented in this paper. The aim is not to necessarily develop a new technology but to evaluate the feasibility and the adaptability of existing ones for our purpose, even if those are not especially dedicated for it. Numerical evaluation is firstly conducted then some of the techniques are experimentally tested. Two of them, the sonic throat and the Fourier Transform Infra Red spectroscopy (FTIR), appear to be particularly promising and allow uncertainties of few percents. The first one allows evaluating the mass flow rate of a supercritical reactive mixture. The second one characterizes the chemical composition before injection in the combustion chamber, for mole fraction greater than 5 %. Copyright © 2009 by the American Institute of Aeronautics and Astronautics, Inc.