Search

The performance of Human Activity Recognition (HAR) models, particularly deep neural networks, is highly contingent upon the availability of the massive amount of annotated training data which should be sufficiently labeled. Though, data acquisition and manual annotation in the HAR domain are prohibitively expensive due to skilled human resource requirements in both steps. Hence, domain adaptation techniques have been proposed to adapt the knowledge from the existing source of data. More recently, adversarial transfer learning methods have shown very promising results in image classification, yet limited for sensor-based HAR problems, which are still prone to the unfavorable effects of the imbalanced distribution of samples. This paper presents a novel generic and robust approach for semi-supervised domain adaptation in HAR, which capitalizes on the advantages of the adversarial framework to tackle the shortcomings, by leveraging knowledge from annotated samples exclusively from the source subject and unlabeled ones of the target subject. Extensive subject translation experiments are conducted on three large, middle, and small-size datasets with different levels of imbalance to assess the robustness and effectiveness of the proposed model to the scale as well as imbalance in the data. The results demonstrate the effectiveness of our proposed algorithms over state-of-the-art methods, which led in up to 13%, 4%, and 13% improvement of our high-level activities recognition metrics for Opportunity, LISSI, and PAMAP2 datasets, respectively. The LISSI dataset is the most challenging one owing to its less populated and imbalanced distribution. Compared to the SA-GAN adversarial domain adaptation method, the proposed approach enhances the final classification performance with an average of 7.5% for the three datasets, which emphasizes the effectiveness of micro-mini-batch training.

In the past few years, the topic of localization has received considerable attention in the research community and especially in mobile robotics area (Borenstein, 1996). It consists of estimating the robot’s pose (position, orientation) with respect to its environment from sensor data. Therefore, better sensory data exploitation is required to increase robot’s autonomy. The simplest way to estimate the pose parameters is integration of odometric data which, however, is associated with unbounded errors, resulting from uneven floors, wheel slippage, limited resolution of encoders, etc. However, such a technique is not reliable due to cumulative errors occurring over the long run. Therefore, a mobile robot must also be able to localize or estimate its parameters with respect to the internal world model by using the information obtained with its external sensors. In system localization, the use of sensory data from a range of disparate multiple sensors, is to automatically extract the maximum amount of information possible about the sensed environment under all operating conditions. Usually, for many problems like obstacle detection, localization or Simultaneous Localization and Map Building (SLAM) (Montemerlo et al., 2002), the perception system of a mobile robot relies on the fusion of several kinds of sensors like video cameras, radars, dead-reckoning sensors, etc. The multi-sensor fusion problem is popularly described by state space equations defining the interesting state, the evolution and observation models. Based on this state space description, the state estimation problem can be formulated as a state tracking problem. To deal with this state observation problem, when uncertainty occurs, the probabilistic Bayesian approaches are the most used in robotics, even if new approaches like the set-membership one (Gning & Bonnifait, 2005) or Belief theory (Ristic and Smets, 2004) have proved themselves in some applications. SLAM is technique used by mobile robots to build up a map within an unknown environment while at the same time keeping track of their current position. Several works implementing SLAM algorithms have been studied extensively over the last years in this direction, leading to approaches that can be classified into three well differentiated paradigms depending on the underlying map structure: metric (Sim et al., 2006) (Tardos et

International audience; Human gait patterns have been intensively studied, both from medical and engineering perspectives, to understand and compensate pathologies. However, the muscle-force sharing problem is still debated as acquiring individual muscle force measurements is challenging, requiring the use of invasive devices.Recent studies, using various objective functions, suggest muscleforce sharing may result from an optimization process. This study proposes using inverse optimal control to identify an objective function. Two popular methods of inverse optimal control, bilevel and inverse Karush-Kuhn-Tucker, were investigated. The identifiedobjective functions were then used to predict muscle forces during gait, and their performances were compared to an exhaustive list of biological cost functions from the literature. The best predictionwas achieved by the bilevel inverse optimal control method, with a root-mean-squared error of 176 N (162 N) and a correlationcoefficient of 0.76 (0.68) for the stance (swing) phase of the gait cycle.These muscle force predictions were thereafter used to compute joint stiffness, exhibiting an average root-mean-square error of 42 Nm.rad?1 and a correlation coefficient of 0.90 when comparedto the reference. The bilevel method's prevalence in terms of robustnessover inverse Karush-Kuhn-Tucker was demonstrated on human data and explained on a toy example.

Purpose Assistive technology products are designed to provide additional accessibility to individuals who have physical or cognitive difficulties, impairments and disabilities. The purpose of this paper is to deal with the control of a knee joint orthosis intended to be used for rehabilitation and assistive purpose; this control aims to reduce the influence of the uncertainties and eliminating the external disturbances in the system. Design/methodology/approach This paper deals with the robust adaptive sliding mode controller (ASMC) of human-driven knee joint orthosis system with mismatched uncertainties and external disturbances. The shank-orthosis system has been modeled and its parameters have been identified. This control reduces the effect of parameter uncertainties and external disturbances on the system performance and improves the system robustness as results. The ASMC was designed to offer the possibility to track the state of the reference model. Moreover, the Lyapunov stability theory was used to study the asymptotical stability of the ASMC. Findings The advantage of the robust ASMC method is the tracking precision and reducing the required time for eliminating external disturbances and uncertainties. The experimental results show in real-time in terms of stability and present that the advantages of this control approach are the position tracking and robustness. Originality/value In this paper, to deal with the parameter uncertainties of the human-driven knee joint orthosis, an ASMC was successfully applied based on sliding mode and Lyapunov stability theory. It has good dynamic response and tracking performance. Besides, the adaptive algorithm is simple, easy to achieve and has good adaptability and robustness against the parameter variations and external disturbances. The design technique is simple and efficient. The development of this control takes into consideration the perturbation, allowing to track a desired trajectory.

International audience; This article describes a new clustering method for segmentation of Magnetic resonance imaging (MRI) brain images. Currently, when fuzzy clustering is applied to brain image segmentation, there are two main problems to be solved which are: (i) the sensitivity to noise and intensity non-uniformity (INU) artifact; (ii) the trapping into local minima and dependency on initial clustering centroids. For the purpose of obtaining satisfactory segmentation performance and dealing with the problems mentioned above, an effective method is developed within the scope of this paper. Firstly, a new objective function utilizing kernelized fuzzy entropy clustering with local spatial information and bias correction (KFECSB) is designed. We then propose a new algorithm based on an improved particle swarm optimization (PSO) with the new fitness function to better segment MRI brain images. To test its performance, the proposed algorithm has been evaluated on several benchmark images including the simulated MRI brain images from the McConnell Brain Imaging Center (BrainWeb) and the real MRI brain images from the Internet Brain Segmentation Repository (IBSR). In addition, a systematic comparison of the proposed algorithm versus five other state of the art techniques is presented. Experimental results show that the proposed algorithm can achieve satisfactory performance for images with noise and intensity inhomogeneity, and provide better results than its competitors.

International audience; This paper discusses a novel conceptual formulation of the fractional-order variational framework for retinex, which is a fractional-order partial differential equation (FPDE) formulation of retinex for the multi-scale nonlocal contrast enhancement with texture preserving. The well-known shortcomings of traditional integer-order computation-based contrast-enhancement algorithms, such as ringing artefacts and staircase effects, are still in great need of special research attention. Fractional calculus has potentially received prominence in applications in the domain of signal processing and image processing mainly because of its strengths like long-term memory, nonlocality, and weak singularity, and because of the ability of a fractional differential to enhance the complex textural details of an image in a nonlinear manner. Therefore, in an attempt to address the aforementioned problems associated with traditional integer-order computation-based contrast-enhancement algorithms, we have studied here, as an interesting theoretical problem, whether it will be possible to hybridize the capabilities of preserving the edges and the textural details of fractional calculus with texture image multi-scale nonlocal contrast enhancement. Motivated by this need, in this paper, we introduce a novel conceptual formulation of the fractional-order variational framework for retinex. First, we implement the FPDE by means of the fractional-order steepest descent method. Second, we discuss the implementation of the restrictive fractional-order optimization algorithm and the fractional-order Courant-Friedrichs-Lewy condition. Third, we perform experiments to analyze the capability of the FPDE to preserve edges and textural details, while enhancing the contrast. The capability of the FPDE to preserve edges and textural details is a fundamental important advantage, which makes our proposed algorithm superior to the traditional integer-order computation-based contrast enhancement algorithms, especially for images rich in textural details.

International audience; Reports on the Robotics and Automation Society (RAS) Technical Committee"s (TC) approval on Wearable Robotics during the IEEE RAS Technical Activities Board meeting that was held 15 May 2016 in Stockholm, Sweden, in conjunction with the IEEE International Conference on Robotics and Automation (ICRA) 2016.

This paper proposes an enhanced Particle Swarm Optimisation (PSO) algorithm and examines its performance. In the proposed PSO approach, PSO is combined with Evolutionary Game Theory to improve convergence. One of the main challenges of such stochastic optimisation algorithms is the difficulty in the theoretical analysis of the convergence and performance. Therefore, this paper analytically investigates the convergence and performance of the proposed PSO algorithm. The analysis results show that convergence speed of the proposed PSO is superior to that of the Standard PSO approach. This paper also develops another algorithm combining the proposed PSO with the Standard PSO algorithm to mitigate the potential premature convergence issue in the proposed PSO algorithm. The combined approach consists of two types of particles, one follows Standard PSO and the other follows the proposed PSO. This enables exploitation of both diversification of the particles’ exploration and adaptation of the search direction.

International audience; This paper presents an adaptive knee-joint based Functional Electrical Stimulation (FES) method to correct the foot drop of paretic patients during swing phase. The rationale behind the adaptive FES is to amplify dorsiflexor stimulation in the late swing when it is most needed in order to facethe increased plantar flexor co-contraction as gastrocnemius muscles are stretched by knee re-extension. To accurately detectthe swing phase (i.e., toes off (TO) and initial contact (IC)), a novel algorithm is proposed by using a foot-mounted inertial measurement unit (IMU). The proposed strategy is verified by experiments conducted with three healthy subjects and three paretic patients. The experimental results show that highly accurate detection of TO/IC can be achieved under different walking speeds and foot contact conditions (normal and abnormalgaits). The clinical experimental results with paretic patients also reveal that similar effects on ankle dorsiflexion can be observed during mid and late swing using the proposed adaptive FES with respect to the classical FES method, while the adaptive FES used lower stimulation intensity.

International audience; An offloading scheme based on LTE-D2D is proposed in this paper to route the intracellular multicast traffic via a network of D2D-enabled User Equipments (UEs). The latter are ready to cooperate under the control of the eNodeB to carry and deliver the traffic. In doing so, the UEs reuse uplink resources granted by the eNodeB and thus, increasing the overall spectral efficiency while reducing the traffic load on the eNodeB. In this paper, we address the joint multicast routing and OFDM resource allocation problem in the D2D network to accomplish the offloading task. To do so, first we formulate the problem as an Integer Linear Programming (ILP) model which takes into account factors that limit spectrum reuse in addition to other LTE-D2D limitations: half-duplex operation and contiguity in resource block allocations. Then, we propose a novel scheme named Joint Multicast Routing and Wireless allocation in D2D communications (JRW-D2D-MC). The devised scheme consists of two-stage algorithm which, first, performs a pre-admittance filtering of flows that can be routed considering the current state of the network. Then, it makes use of the branch-and-cut method to solve the reduced ILP model. To evaluate effectiveness of our proposal, we implement the LTE-D2D standard in a network simulator NS-3. The results are very good in terms of flow-acceptance rate and latency.

International audience; The emerging and fastest growing multimedia services such as Skype, GTalk and interactive video gaming have created new challenges for wireless communication technologies, especially in terms of resource allocation and power optimization of User Equipments (UEs). The efficient resources and the power optimization are very important in the next communication systems because new multimedia services are more resource and power hungry. Having more traffic flow in the downlink as compared to the uplink, the resource allocation schemes in the downlink are more important than those in the uplink. In this paper, we have proposed a new downlink scheduling scheme, known as Quality-aware DRX (Q-DRX) Scheme, for Long Term Evaluation (LTE) wireless networks that not only improves the Quality of Service (QoS) to satisfy the user׳s experience of multimedia services but also saves the UE׳s power by using the Discontinuous Reception (DRX). This paper investigates how the different durations of DRX Light and Deep Sleep cycles affect the QoS and influence the Quality of Experience (QoE) of end users, in the context of Voice over IP (VoIP) traffic delivery over the LTE wireless network. The performance of the Q-DRX scheme is evaluated with traditional scheduling schemes, in the perception of the following important parameters, System Throughput, Throughput Fairness Index, Packet Loss Rate and Packet Delay. Our proposed Q-DRX Scheme successfully minimized the packet׳s delay and packet׳s loss by considering the key parameters of QoS with fair resources distribution among the UEs to achieve a high satisfaction level.

International audience; The decoupling of control and data planes in Software-Defined Networking (SDN) brings benefits in terms of logically centralized control and application programming. But, the single point of management in physically centralized SDN architectures is a potential point of failure and a bottleneck that compromises network reliability and performance. Such centralized designs may also face scalability challenges especially in networks with a large number of hosts (e.g. IoT-like networks). To avoid such concerns, SDN control architectures are usually designed as physically distributed systems. This raises practical challenges about the best approach to decentralizing the control plane while maintaining the logically centralized network view. In particular, determining the number of controllers and locating them in the network is a hard task that should be addressed appropriately. This paper proposes two novel strategies that cover different aspects of the controller placement problem with respect to performance and reliability criteria. These strategies use two types of heuristics that are compared and assessed on large-scale topologies to provide operators with guidelines on how to find their optimal controller placement that meets their specific needs.

International audience; Creating a standard for knowledge representation and reasoning in autonomous robotics is an urgent task if we consider recent advances in robotics as well as predictions about the insertion of robots in human daily life. Indeed, this will impact the way information is exchanged between multiple robots or between robots and humans and how they can all understand it without ambiguity. Indeed, Human Robot Interaction (HRI) represents the interaction of at least two cognition models (Human and Robot). Such interaction informs task composition, task assignment, communication, cooperation and coordination in a dynamic environment, requiring a flexible representation. Hence, this paper presents the IEEE RAS Autonomous Robotics (AuR) Study Group, which is a spin-off of the IEEE Ontologies for Robotics and Automation (ORA) Working Group, and and its ongoing work to develop the first IEEE-RAS ontology standard for autonomous robotics. In particular, this paper reports on the current version of the ontology for autonomous robotics as well as on its first implementation successfully validated for a human-robot interaction scenario, demonstrating the developed ontology's strengths which include semantic interoperability and capability to relate ontologies from different fields for knowledge sharing and interactions.

Medical images, such as Computed Axial Tomography (CAT), Magnetic Resonance Imaging (MRI), Ultrasound, and X-Ray, in standard DICOM (Digital Imaging and Communications in Medicine) formats are often stored in Picture Archiving and Communication Systems (PACS) and linked with other clinical information in clinical management systems.

With the rise of Video-on-Demand (VoD) systems as a preferred way to distribute video content over IP networks, many research works and innovations have focused on improving the scalability of streaming systems by looking at distributed approaches such as peer-to-peer (P2P). One of the most critical aspects in P2P-assisted streaming system is the real-time resource allocation, which drives the performance of the system in terms of capacity utilization and VoD requests rejection rates. In this paper, we specifically focus on the problem of maximizing the P2P streaming system utilization by effectively alternating between different resource allocation strategies. Switching between different resource allocation strategies is guided by a run-time statistical analysis of performances against predicted content popularity pattern. A key contribution of this paper resides in effectively combining different, and potentially conflicting, performance objectives when deciding on which resource allocation strategy to use. Indeed, we use a Bayesian Fusion to select the most appropriate resource allocation strategy to deal with future content demand. With our P2P resource allocation framework, a VoD service operator can combine any number of resource allocation strategies and formulate different performance objectives that meet the requirements of its network and the content consumption behavior of its users.

In this paper, an innovative watermarking scheme based on differential evolution (DE) in the transform domain is proposed. The insertion and extraction of the watermark are performed in discrete wavelet transform-singular value decomposition (DWT–SVD) transform domain. In the embedding process, the host image is transformed into sub-bands of different frequencies by third-level DWT and then subsequent application of SVD on low pass (LL) and high pass (HH) sub-bands at level third. The watermark image is properly scaled down by multiplying with different scaling factors (SFs) and embedded in the Singular value matrix of LL and HH sub-bands of the host image to make the watermark invisible and robust. We applied an optimization technique, differential evolution, to search optimal scaling factors to improve the quality of watermarked image and robustness of the watermark. In order to overcome the false positive problem, a binary watermark is also embedded in the host image in a lossless manner. According to the numerical results, the quality of watermarked image is satisfactory and embedded watermark is extracted successfully even if the watermarked image is exposed to various image processing and geometric attacks.

International audience; As an important movement of the daily living activities, sit-to-stand (STS) movement is usually a difficult task facing elderly and dependent people. To provide appropriate power assistance for the sit-to-stand movement, a novel intention-based Active Impedance Control (AIC) strategy applied on a lower limb exoskeleton is proposed in this paper. The AIC is able to adapt the mechanical impedance of the human-exoskeleton system towards a desired one using the exoskeleton's power assistance. In the AIC structure, a human joint torque observer is designed to estimate the human joint torques using joint angles information instead of electromyography (EMG) or force/torque sensors; a time-varying desired impedance model is proposed according the wearer's lower limb motion ability. Simulations were implemented to illustrate the characteristics and performances of the proposed approach. Experiments with a healthy subject were carried out to evaluate the effectiveness of the proposed method. The experiments show satisfactory results in terms of appropriate power assist based on the wearer's motion intention.

International audience; We address the online virtual network embedding problem within the Cloud's backbone to optimally map the virtual routers and links in the substrate network in order to maximize the Cloud's provider revenue. Since the problem is NP-hard, we propose a novel approach, named VNE-GH, to significantly reduce the problem size using the Gomory-Hu transformation without losing useful information on the virtual network embedding problem. Starting from the Gomory-Hu compact tree structure, we formulate the virtual network embedding as an Integer Linear Program and resolve the reduced size problem using the branch- and-cut algorithm. Results obtained via extensive simulations show that VNE-GH outperforms the most prominent related work strategies in terms of i) acceptance rate of virtual network requests and ii) Cloud provider's revenue.

International audience; no abstract

In this study, the authors present a new image segmentation algorithm based on two-dimensional digital fractional integration (2D-DFI) that was inspired from the properties of the fractional integration function. Although obtaining a good segmentation result corresponds to finding the optimal 2D-DFI order, the authors propose a new alternative based on Legendre moments. This framework, called two dimensional digital fractional integration and Legendre moments’ (2D-DFILM), allows one to include contextual information such as the global object shape and exploits the properties of the 2D fractional integration. The efficiency of 2D-DFILM is shown by the comparison to other six competing methods recently published and it was tested on real-world problem.

Service composition in ubiquitous and pervasive environments is becoming an active research domain which has received widespread attention in recent years. It aims to offer seamless access to a variety of high level and complex functionalities by combining existing services. Several frameworks have been designed to support service composition in ubiquitous and pervasive environments. Although some ubiquitous requirements and challenges are relatively well addressed by the proposed frameworks, others are still at a preliminary stage and should be well explored such as, automatic service composition with little human intervention, context and quality of service management, and service selection under uncertainty and changes. For this end, we propose in this paper a layered design approach for flexible and failure tolerant service composition using two main phases: off-line phase and on-line phase. In the off-line phase, a global graph that links all the available abstract services is generated automatically using rule-based technique. The defined rules aim at optimizing both the number of services and parameters that appear in the global graph. In the on-line phase, a subgraph is extracted spontaneously from the global graph according to the occurred and detected event in the environment at real time. Thereafter, the extracted subgraph is performed using service selection strategies. A prototype implementation including real services for event detection in smart home shows clearly the feasibility of the proposed approach in real environment. Also, the set of performed evaluation tests reveals the interest and the performance of the proposed algorithms.

In this paper, a hybrid moment/position controller in task space is proposed for tasks involving a contact between a robot and its environment. We consider a contour-tracking task performed by a six DOF (Degrees Of Freedom) parallel robot. The task space dynamic model of the robot in contact with its environment, seen as a black box, is estimated by a MLP-NN (MultiLayer Perceptron Neural Network). The neural network non-linearity is treated using Taylor series expansion. An adaptation algorithm of the neural parameters resulting from a closed-loop stability analysis is proposed. The performance of the proposed controller is validated on the C5 parallel robot by considering two different environments: rigid and compliant.

International audience; The wide acceptance of wheelchairs in the society may be associated to their simplicity, and to the important mobility ease they provide to people with disability. Nonetheless, users with extreme mobility impairments still experience some cognitive and physical steering challenges that limit their capacity to navigate even simple environments. Assistive wheelchairs with functionalities to enable customisation of user handicap are therefore still required. The diversity of user disabilities and needs, and the corresponding assistive measures, necessitate proper understanding of the user. Modelling and integration of user driving behaviour into the control loop in order to monitor user inputs, and execute necessary corrective steps in troublesome situations, may just be one way of providing the assistance that corresponds to needs of the user. This paper attempts to identify the steering behaviour of an individual wheelchair user in a normal living room environment. The preliminary navigation experiments required to obtain the steering data are performed in a virtual world on an augmented wheelchair platform, and the participant is observed to exhibit expected navigational behaviours. Least square method is considered in the estimation of model parameters for the non-linear steering model with interesting regression analysis outcome.

International audience; In recent years, much attention has been directed to building robots with hybrid and multi-modal locomotion in air, water and on the ground. This rise in interest is due to the enormous requirements of multi-domain earthquake rescue, pollution monitoring, natural species discovery and other applications in which multi-modal locomotion can offer unprecedented advantages to robot mobility. Swimming, crawling, rolling, walking, running, jumping and flying are quite common in the animal kingdom, and such locomotion occurs in very different physical environments. Interestingly, an animal may perform two or more modes of locomotion. For instance, turtles and salamanders can swim underwater and walk on land, while swans can swim and fly effectively. Their often remarkable abilities could inspire innovative designs to improve the way man-made systems operate in and interact with challenging outdoor environments consisting of multiple media. Modes of locomotion in nature can be decomposed broadly into three categories: terrestrial, aerial and aquatic. There are situations in which the distinction between categories is vague, such as movement on the surface of water or in underground tunneling. However, generally, the types of locomotion used by animals can be categorized into one of these three areas. The primary venue of locomotion for many animals tends to be mono-modal; the majority of their task-space demands only a singular modality, with minimal need of alternative morphologies for transportation. The fundamental reason for multi-modal animal locomotion is for survival. The need for these multiple modes can arise from different requirements related to survival, including fast escape, fast pursuit, searching for food, breeding, nesting, conserving energy and migration. The paper presents a brief survey on the recent application works of hybrid and multi-modal locomotion.

International audience; Most range-free techniques for indoor localization depend on the received signal strength (RSS) fingerprints. Their performances are relied to the structure of the considered indoor environments. We consider in this paper RSS-based methods: Multi-Layer Perceptron Neural Network (MLPNN), and K-nearest neighbor (KNN), and compare their performance under the same indoor environment. One of the advantages focused by the choice of these techniques is their robustness against external disturbances that may affect the received RSS signal. Moreover, we propose a new metric to enhance the performance of the KNN method, called d-nearest neighbor. In order to test the different techniques, we build a heterogeneous fingerprint database with different resolutions. The obtained results show the efficiency of the proposed enhancement in the case of a heterogeneous high resolution database.

International audience; This paper presents a new approach to identify persons using their fingerprints. The proposed approach is based on the Polynomial Discrete Radon Transform (PDRT). The PDRT extends the Classical Radon Transform (RT) by generalizing the curves used for the projection. In fact, the classical Radon projection is restricted to straight lines. However, the PDRT method projects the image along polynomial curves following different directions. The proposed technique enables the extraction of directional characteristics from the fingerprint of the hand. Since several methods in the literature reached a level of success on fingerprint recognition, this paper proposes a new identification method which is invariant to geometrical transformations such as translation and rotation. In addition, it is robust to noise. To evaluate the performance of this approach, we have tested it on a database containing 128 fingerprint images. The obtained result confirms the efficiency of the method.

This paper proposes to reduce the computational time of an algorithm based on the combination of the Evolutionary Game Theory (EGT) and the Particle Swarm Optimisation (PSO), named C-EGPSO, by using Neural Networks (NN) in order to lighten the computation of the identified heavy part of the C-EGPSO. This computationally burdensome task is the resolution of the EGT part that consists in solving iteratively a differential equation in order to optimally adapt the direction search and the size step of the PSO at each iteration. Therefore, it is proposed to use NN to learn the solution of this differential equation according to the initial conditions in order to gain a precious time.

International audience; no abstract

In image processing, finding the optimal threshold(s) for an image with a multimodal histogram can be done by solving a Gaussian curve fitting problem, i.e. fitting a sum of Gaussian probability density functions to the image histogram. This problem can be expressed as a continuous nonlinear optimization problem. The goal of this paper is to show the relevance of using a recently proposed variant of the Particle Swarm Optimization (PSO) algorithm, called PSO-2S, to solve this image thresholding problem. PSO-2S is a multi-swarm PSO algorithm using charged particles in a partitioned search space for continuous optimization problems. The performances of PSO-2S are compared with those of SPSO-07 (Standard Particle Swarm Optimization in its 2007 version), using reference images, i.e. using test images commonly used in the literature on image segmentation, and test images generated from brain MRI simulations. The experimental results show that PSO-2S produces better results than SPSO-07 and improves significantly the stability of the segmentation method.

International audience; no abstract

In recent years dynamic optimization problems have attracted a growing interest from the community of stochastic optimization researchers with several approaches developed to address these problems. The goal of this chapter is to present the different tools and benchmarks to evaluate the performances of the proposed algorithms. Indeed, testing and comparing the performances of a new algorithm to the different competing algorithms is an important and hard step in the development process. The existence of benchmarks facilitates this step, however, the success of these benchmarks is conditioned by their use by the community. In this chapter, we cite many tested problems (we focused only on the continuous case), and we only present the most used: the moving peaks benchmark , and the last proposed: the generalized approach to construct benchmark problems for dynamic optimization (also called benchmark GDBG).

Wearable robots have defined a new horizon for elderly and disabled people, to regain control of their limbs, as well as for healthy people, to increase their abilities of hard missions execution. The present paper deals with the control of a lower limb orthosis applied at the knee joint level for rehabilitation purposes. A bounded control torque is developed in order to guarantee the asymptotic stability of the knee orthosis. The control law respects the physical constraints of the system. Moreover, it is robust with respect to external disturbances. The effectiveness of the control torque is tested in real-time using the EICOSI orthosis of the LISSI Lab.

Enabling cross-organizational cooperation in ubiquitous computing environments poses new security challenges that concern particularly the interoperability of security management systems and the security policies of each organization. In this paper, we present a semantic framework for cooperative security management processes design in a cross-organizational context. Our framework is based on a hybrid approach that caters between the advantages of bottom-up and top down approaches. The cooperation model of our framework is based on the composition of atomic security management processes, by using speech acts and ontologies, and their mapping with internal processes views. The establishment of an e-contract between partners allows specifying common terminology for exchanging messages, cooperation security policy and process control flows. A scenario of cooperative process is presented to demonstrate the feasibility of the proposed framework.

International audience; This paper presents an analysis of stabilogram using the Principal Component Analysis (PCA) decomposition. It shows also the effects of different aspects on the human postural stability. The stabilogram measures either lateral displacement or forward-backward displacement of a subject. These measurements are taken to quantify posture while standing in one of four controlled positions. By using Principal Component Analysis (PCA), the stabilogram is decomposed into three components with biological meaning. The components are trend, rambling and trembling. This paper proposes to create analytic signals for rambling (deterministic) and trembling (random) and use the resulting complex trajectories to identify the effect of age and gender on postural stability. The proposed method employs a signal analysis front end (PCA analysis) and a signal interpretation backend (clustering of complex trajectories). Experimental results show the efficiency of the PCA analysis to identify the effect of age and gender on the postural stability. These results are able to discriminate between control and young groups and indicate a less well-controlled posture for control subjects (34.5 ± 7.5yrs) relatively to young subjects (22.5 ± 2.5yrs). Results are also able to discriminate between female subjects and male subjects and indicate a less well-controlled posture for female subjects relatively to males.

This brief paper deals with using the Particle Swarm Optimization metaheuristic for optimally sizing switched current (SI) memory cells, namely the class AB grounded gate SI memory cell. Pareto front is generated while optimizing two main conflicting performances: maximizing both the signal to noise ratio and the sampling frequency. SPICE simulation results are presented to validate obtained sizing.

This chapter addresses to the blocking flowshop scheduling problem with the aim of minimizing the makespan. An Estimation of Distribution Algorithm, followed by a local search procedure, after the step of creating a new individual, was developed in order to solve this problem. Our comparisons were performed against representative approaches proposed in the literature related to the blocking flowshop scheduling problem. The obtained results have shown that the proposed algorithm is able to improve 109 out of 120 best known solutions of Taillard’s instances. Moreover, our algorithm outperforms all competing approaches in terms of solution quality and computational time.

Most of the works that address the flowshop scheduling problem presume unlimited buffers between successive machines. However, with the advent of new technologies in the manufacturing systems, limited capacity storage between machines has become profitable. Aimed at makespan minimization, the flowshop scheduling problem with buffer constraints is NP-hard in the strong sense. Therefore, several approximate algorithms have been proposed in the literature. In this chapter, we propose an Estimation of Distribution Algorithm for solving a flowshop scheduling problem with buffer constraints. The main characteristics of the problem, such as the order of jobs and similar blocks of jobs in the sequence, are taken into account while building the probabilistic model. In order to enrich the search procedure of the algorithm, a skewed variable neighbourhood search algorithm is embedded into it, restricted by a calculated probability which depends on the quality of the created offspring. The computational results show that our algorithm outperforms genetic algorithm and particle swarm algorithm, and can obtain several optimal solutions in a short time.

This paper presents a method for the assessment of left ventricular myocardial deformations from untagged cardiac cine MRI. Firstly, a level set segmentation process is applied on the dataset to detect both endocardial and epicardial boundaries. Then the successive contours are matched using an original procedure that consists in an alignment followed by a morphing process. From the matched contours, we deduced an initial velocity of the contour points that will be used in a Thin-Plate Splines approximation method for estimating of the velocity flow on the complete myocardial structure. Finally, local measurements of ventricular deformations are derived from the velocity flow. The validation of the method is performed both mathematically and by comparing the measurements to those obtained on the same patients with the HARmonic Phase reference (HARP) method applied on matched tagged MR images.

International audience; Assistive and rehabilitation robotics currently covers a broad spectrum of research areas, from intelligent robots that act as aides, companions and educators; to robots that provide physical assistance for mobility, manipulation and support; to embedded robotics, ambient intelligence and intelligent spaces. People whose ability to perform daily living activities is limited or inhibited by injury, disease or impairment include (but are not limited to) stroke survivors with sustained neurological injuries, elderly populations with skeletal muscle weakness, persons with low-vision or blindness, traumatic injuries to the brain or spinal cord, and a variety of degenerative diseases that affect cognition, muscle tone and motor control.Over the last decade, robotics technologies increasingly interact closely with the humans they are assisting. Improvements in sensor technologies, hardware design and control have enabled this. Augmenting assistive machines with robotics intelligence in particular has made great strides—including intelligence found in wearable robots, prostheses, intelligent walkers and canes, and autonomy found in wheelchairs and robotic arms. Moreover, the emergence of novel adapted technologies, such as those in wearable sensors and soft robots, are gaining in popularity—and not the least because of considerable reductions in cost, size and energy consumption. There also has been a considerable push for assistive robots that serve as cognitive and educational aides, and provide coaching to improve health outcomes. Importantly, subject studies with such robots performed in-situ—in the human’s environment rather than the laboratory—have been gaining momentum.This Special Issue on Assistive and Rehabilitation Robotics grew out of the joint efforts of three distinct workshops that were organized in conjunction with the 2014 IEEE/RSJ International Conference on Robotics and Intelligent Systems (IROS), held in Chicago, Illinois USA, from September 14th to 18th, 2014. These workshops were entitled: “Assistance and Service Robotics in a Human Environment”,1 “Rehabilitation and Assistive Robotics: Bridging the Gap Between Clinicians and Roboticists”,2 and “Assistive Robots for Individuals with Disabilities: HRI Issues and Beyond”.3The three workshops gathered leading researchers in the field of assistive and rehabilitation robotics. Fruitful discussion among the speakers and the attendees centered around the state of the art and the challenges and limiting factors for developing sustainable robotics solutions for human assistance and rehabilitation.Following the three workshops, we guest editors launched this Special Issue through a call for papers that included the participants of the three workshops and also the larger research community. Following the open call, we collected 98 extended abstracts on topics ranging from human-robot interaction and social interaction, to healthcare and well-being, to intelligent habitats and mobility assistance. After a careful review process, 22 abstracts were invited to submit full papers. Of these, 14 papers were accepted for publication in the Special Issue.The papers of the Special Issue cover a broad range of topics, from robots that assist with manipulation and mobility (Sect. 3.1), to those that aid in education, cognitive and physical rehabilitation using social interaction capabilities (Sect. 3.2), to those that physically augment the human body (Sect. 3.3).

This paper deals with visual evaluation of object distances using Soft-Computing based approaches and pseudo-3D standard low-cost sensor, namely the Kinect. The investigated technique points toward robots’ vision and visual metrology of the robot’s surrounding environment. The objective is providing the robot the ability of evaluating distances between objects in its surrounding environment. In fact, although presenting appealing advantages, the Kinect has not been designed for metrological aims. The investigated approach offers the possibility to use this low-cost pseudo-3D sensor for distance evaluation avoiding 3D feature extraction and thus exploiting the simplicity of only 2D image’ processing. Experimental results show the viability of the proposed approach and provide comparison between different machine learning techniques as Adaptive-network-based fuzzy inference (ANFIS), Multi-layer Perceptron (MLP), Support vector regression (SVR), Bilinear interpolation.

In this work we contribute to development of a real-time human-like intuitive artificial vision system taking advantage from visual attention skill. Implemented on a 6-wheels mobile robot equipped with communication facilities, such a system allows detecting combustion perimeter in real outdoor environment without prior knowledge. It opens appealing perspectives in fire-fighting strategy enhancement and in early-stage woodland fire’s detection. We provide experimental results showing as well the plausibility as the efficiency of the proposed system.