scispace - formally typeset
Search or ask a question

Showing papers on "Robustness (computer science) published in 2010"


Proceedings ArticleDOI
13 Jun 2010
TL;DR: A new type of correlation filter is presented, a Minimum Output Sum of Squared Error (MOSSE) filter, which produces stable correlation filters when initialized using a single frame, which enables the tracker to pause and resume where it left off when the object reappears.
Abstract: Although not commonly used, correlation filters can track complex objects through rotations, occlusions and other distractions at over 20 times the rate of current state-of-the-art techniques. The oldest and simplest correlation filters use simple templates and generally fail when applied to tracking. More modern approaches such as ASEF and UMACE perform better, but their training needs are poorly suited to tracking. Visual tracking requires robust filters to be trained from a single frame and dynamically adapted as the appearance of the target object changes. This paper presents a new type of correlation filter, a Minimum Output Sum of Squared Error (MOSSE) filter, which produces stable correlation filters when initialized using a single frame. A tracker based upon MOSSE filters is robust to variations in lighting, scale, pose, and nonrigid deformations while operating at 669 frames per second. Occlusion is detected based upon the peak-to-sidelobe ratio, which enables the tracker to pause and resume where it left off when the object reappears.

2,948 citations


Journal ArticleDOI
TL;DR: In this article, the authors introduce a new class of structured compressible signals along with a new sufficient condition for robust structured compressibility signal recovery that they dub the restricted amplification property, which is the natural counterpart to the restricted isometry property of conventional CS.
Abstract: Compressive sensing (CS) is an alternative to Shannon/Nyquist sampling for the acquisition of sparse or compressible signals that can be well approximated by just K ? N elements from an N -dimensional basis. Instead of taking periodic samples, CS measures inner products with M < N random vectors and then recovers the signal via a sparsity-seeking optimization or greedy algorithm. Standard CS dictates that robust signal recovery is possible from M = O(K log(N/K)) measurements. It is possible to substantially decrease M without sacrificing robustness by leveraging more realistic signal models that go beyond simple sparsity and compressibility by including structural dependencies between the values and locations of the signal coefficients. This paper introduces a model-based CS theory that parallels the conventional theory and provides concrete guidelines on how to create model-based recovery algorithms with provable performance guarantees. A highlight is the introduction of a new class of structured compressible signals along with a new sufficient condition for robust structured compressible signal recovery that we dub the restricted amplification property, which is the natural counterpart to the restricted isometry property of conventional CS. Two examples integrate two relevant signal models-wavelet trees and block sparsity-into two state-of-the-art CS recovery algorithms and prove that they offer robust recovery from just M = O(K) measurements. Extensive numerical simulations demonstrate the validity and applicability of our new theory and algorithms.

1,789 citations


Proceedings ArticleDOI
13 Jun 2010
TL;DR: A novel image operator is presented that seeks to find the value of stroke width for each image pixel, and its use on the task of text detection in natural images is demonstrated.
Abstract: We present a novel image operator that seeks to find the value of stroke width for each image pixel, and demonstrate its use on the task of text detection in natural images. The suggested operator is local and data dependent, which makes it fast and robust enough to eliminate the need for multi-scale computation or scanning windows. Extensive testing shows that the suggested scheme outperforms the latest published algorithms. Its simplicity allows the algorithm to detect texts in many fonts and languages.

1,531 citations


Journal ArticleDOI
TL;DR: This paper presents a method based on robust statistics to register images in the presence of differences, such as jaw movement, differential MR distortions and true anatomical change, which is highly accurate and shows superior robustness with respect to noise, to intensity scaling and outliers.

1,132 citations


Journal ArticleDOI
TL;DR: This work motivates and proposes new versions of the diffusion LMS algorithm that outperform previous solutions, and provides performance and convergence analysis of the proposed algorithms, together with simulation results comparing with existing techniques.
Abstract: We consider the problem of distributed estimation, where a set of nodes is required to collectively estimate some parameter of interest from noisy measurements. The problem is useful in several contexts including wireless and sensor networks, where scalability, robustness, and low power consumption are desirable features. Diffusion cooperation schemes have been shown to provide good performance, robustness to node and link failure, and are amenable to distributed implementations. In this work we focus on diffusion-based adaptive solutions of the LMS type. We motivate and propose new versions of the diffusion LMS algorithm that outperform previous solutions. We provide performance and convergence analysis of the proposed algorithms, together with simulation results comparing with existing techniques. We also discuss optimization schemes to design the diffusion LMS weights.

1,116 citations


Journal ArticleDOI
TL;DR: The proposed OP-ELM methodology performs several orders of magnitude faster than the other algorithms used in this brief, except the original ELM, and is still able to maintain an accuracy that is comparable to the performance of the SVM.
Abstract: In this brief, the optimally pruned extreme learning machine (OP-ELM) methodology is presented. It is based on the original extreme learning machine (ELM) algorithm with additional steps to make it more robust and generic. The whole methodology is presented in detail and then applied to several regression and classification problems. Results for both computational time and accuracy (mean square error) are compared to the original ELM and to three other widely used methodologies: multilayer perceptron (MLP), support vector machine (SVM), and Gaussian process (GP). As the experiments for both regression and classification illustrate, the proposed OP-ELM methodology performs several orders of magnitude faster than the other algorithms used in this brief, except the original ELM. Despite the simplicity and fast performance, the OP-ELM is still able to maintain an accuracy that is comparable to the performance of the SVM. A toolbox for the OP-ELM is publicly available online.

745 citations


Book ChapterDOI
08 Sep 2010
TL;DR: This work considers temporal logic formulae specifying constraints in continuous time and space on the behaviors of continuous and hybrid dynamical system admitting uncertain parameters and presents several variants of robustness measures that indicate how far a given trajectory stands, in space and time, from satisfying or violating a property.
Abstract: We consider temporal logic formulae specifying constraints in continuous time and space on the behaviors of continuous and hybrid dynamical system admitting uncertain parameters. We present several variants of robustness measures that indicate how far a given trajectory stands, in space and time, from satisfying or violating a property. We present a method to compute these robustness measures as well as their sensitivity to the parameters of the system or parameters appearing in the formula. Combined with an appropriate strategy for exploring the parameter space, this technique can be used to guide simulation-based verification of complex nonlinear and hybrid systems against temporal properties. Our methodology can be used for other non-traditional applications of temporal logic such as characterizing subsets of the parameter space for which a system is guaranteed to satisfy a formula with a desired robustness degree.

661 citations


Journal ArticleDOI
TL;DR: An integral predictive and nonlinear robust control strategy to solve the path following problem for a quadrotor helicopter with parametric and structural uncertainties presented to corroborate the effectiveness and the robustness of the proposed strategy.

643 citations


Proceedings ArticleDOI
03 May 2010
TL;DR: This work proposes an extension to this approach to vehicle localization that yields substantial improvements over previous work in vehicle localization, including higher precision, the ability to learn and improve maps over time, and increased robustness to environment changes and dynamic obstacles.
Abstract: Autonomous vehicle navigation in dynamic urban environments requires localization accuracy exceeding that available from GPS-based inertial guidance systems. We have shown previously that GPS, IMU, and LIDAR data can be used to generate a high-resolution infrared remittance ground map that can be subsequently used for localization [4]. We now propose an extension to this approach that yields substantial improvements over previous work in vehicle localization, including higher precision, the ability to learn and improve maps over time, and increased robustness to environment changes and dynamic obstacles. Specifically, we model the environment, instead of as a spatial grid of fixed infrared remittance values, as a probabilistic grid whereby every cell is represented as its own gaussian distribution over remittance values. Subsequently, Bayesian inference is able to preferentially weight parts of the map most likely to be stationary and of consistent angular reflectivity, thereby reducing uncertainty and catastrophic errors. Furthermore, by using offline SLAM to align multiple passes of the same environment, possibly separated in time by days or even months, it is possible to build an increasingly robust understanding of the world that can be then exploited for localization. We validate the effectiveness of our approach by using these algorithms to localize our vehicle against probabilistic maps in various dynamic environments, achieving RMS accuracy in the 10cm-range and thus outperforming previous work. Importantly, this approach has enabled us to autonomously drive our vehicle for hundreds of miles in dense traffic on narrow urban roads which were formerly unnavigable with previous localization methods.

615 citations


Proceedings ArticleDOI
03 May 2010
TL;DR: This paper describes a navigation system that allowed a robot to complete 26.2 miles of autonomous navigation in a real office environment, including an efficient Voxel-based 3D mapping algorithm that explicitly models unknown space.
Abstract: This paper describes a navigation system that allowed a robot to complete 26.2 miles of autonomous navigation in a real office environment. We present the methods required to achieve this level of robustness, including an efficient Voxel-based 3D mapping algorithm that explicitly models unknown space. We also provide an open-source implementation of the algorithms used, as well as simulated environments in which our results can be verified.

536 citations


Proceedings ArticleDOI
01 Dec 2010
TL;DR: This article describes the design of a linear robust dynamic output feedback control scheme for output reference trajectory tracking tasks in a leader-follower non-holonomic car formation problem using the cars' kinematic models.
Abstract: This article describes the design of a linear robust dynamic output feedback control scheme for output reference trajectory tracking tasks in a leader-follower non-holonomic car formation problem using the cars' kinematic models. A simplification is proposed on the follower's exact open loop position tracking error dynamics, obtained by flatness considerations, resulting in a system described by an additively disturbed set of two, second order, integrators with non-linear velocity dependent control input matrix gain. The unknown disturbances are modeled as absolutely bounded, additive, unknown time signals which may be locally approximated by arbitrary elements of, a, fixed, sufficiently high degree family of Taylor polynomials. Linear Luenberger observers may be readily designed, which include the, self updating, internal model of the unknown disturbance input vector components as generic time-polynomial models. The proposed Generalized Proportional Integral (GPI) observers, which are the dual counterpart of GPI controllers ([11]), achieve a, simultaneous, disturbance estimation and tracking error phase variables estimation. This, on-line, gathered information is used to advantage on the follower's linear output feedback controller thus allowing for a simple, yet efficient, disturbance and control input gain cancelation effort. The results are applied to control the fixed time delayed trajectory tracking of the leader path on the part of the follower. Simulations are presented which illustrate the robustness of the proposed approach.

Journal ArticleDOI
TL;DR: Saeys et al. as discussed by the authors proposed a large-scale analysis of ensemble feature selection, where multiple feature selections are combined in order to increase the robustness of the final set of selected features.
Abstract: Motivation: Biomarker discovery is an important topic in biomedical applications of computational biology, including applications such as gene and SNP selection from high-dimensional data. Surprisingly, the stability with respect to sampling variation or robustness of such selection processes has received attention only recently. However, robustness of biomarkers is an important issue, as it may greatly influence subsequent biological validations. In addition, a more robust set of markers may strengthen the confidence of an expert in the results of a selection method. Results: Our first contribution is a general framework for the analysis of the robustness of a biomarker selection algorithm. Secondly, we conducted a large-scale analysis of the recently introduced concept of ensemble feature selection, where multiple feature selections are combined in order to increase the robustness of the final set of selected features. We focus on selection methods that are embedded in the estimation of support vector machines (SVMs). SVMs are powerful classification models that have shown state-of-the-art performance on several diagnosis and prognosis tasks on biological data. Their feature selection extensions also offered good results for gene selection tasks. We show that the robustness of SVMs for biomarker discovery can be substantially increased by using ensemble feature selection techniques, while at the same time improving upon classification performances. The proposed methodology is evaluated on four microarray datasets showing increases of up to almost 30% in robustness of the selected biomarkers, along with an improvement of ~15% in classification performance. The stability improvement with ensemble methods is particularly noticeable for small signature sizes (a few tens of genes), which is most relevant for the design of a diagnosis or prognosis model from a gene signature. Contact: yvan.saeys@psb.ugent.be Supplementary information: Supplementary data are available at Bioinformatics online.

Journal ArticleDOI
TL;DR: In this note, the robust output regulation problem of a multi-agent system is considered and an internal model based distributed control scheme is adopted to achieve the objectives of asymptotic tracking and disturbance rejection in an uncertain multi- agent system.
Abstract: In this note, the robust output regulation problem of a multi-agent system is considered. An internal model based distributed control scheme is adopted to achieve the objectives of asymptotic tracking and disturbance rejection in an uncertain multi-agent system where both the reference inputs and disturbances are generated by an exosystem.

Proceedings ArticleDOI
13 Jun 2010
TL;DR: This work shows that augmenting an on-line learning method with complementary tracking approaches can lead to more stable results, and uses a simple template model as a non-adaptive and thus stable component, a novel optical-flow-based mean-shift tracker as highly adaptive element and anon-line random forest as moderately adaptive appearance-based learner.
Abstract: Tracking-by-detection is increasingly popular in order to tackle the visual tracking problem. Existing adaptive methods suffer from the drifting problem, since they rely on self-updates of an on-line learning method. In contrast to previous work that tackled this problem by employing semi-supervised or multiple-instance learning, we show that augmenting an on-line learning method with complementary tracking approaches can lead to more stable results. In particular, we use a simple template model as a non-adaptive and thus stable component, a novel optical-flow-based mean-shift tracker as highly adaptive element and an on-line random forest as moderately adaptive appearance-based learner. We combine these three trackers in a cascade. All of our components run on GPUs or similar multi-core systems, which allows for real-time performance. We show the superiority of our system over current state-of-the-art tracking methods in several experiments on publicly available data.

Journal ArticleDOI
TL;DR: A sufficient and necessary condition for the robust asymptotical stability of fractional-order interval systems with the fractional order α satisfying 0 < α < 1 is presented.
Abstract: This technical note firstly presents a sufficient and necessary condition for the robust asymptotical stability of fractional-order interval systems with the fractional order α satisfying 0 < α < 1. And then a sufficient condition for the robust asymptotical stabilization of such fractional-order interval systems is derived. All the results are obtained in terms of linear matrix inequalities. Finally, two illustrative examples are given to show that our results are effective for checking the robust stability and designing the robust stabilizing controller for fractional-order interval systems.

Journal ArticleDOI
TL;DR: A new tuning method for fractional order proportional and derivative (PD ¿) or FO-PD controller is proposed for a class of typical second-order plants and shows that the closed-loop system can achieve favorable dynamic performance and robustness.
Abstract: In recent years, it is remarkable to see the increasing number of studies related to the theory and application of fractional order controller (FOC), specially PI ? D ? controller, in many areas of science and engineering. Research activities are focused on developing new analysis and design methods for fractional order controllers as an extension of classical control theory. In this paper, a new tuning method for fractional order proportional and derivative (PD ?) or FO-PD controller is proposed for a class of typical second-order plants. The tuned FO-PD controller can ensure that the given gain crossover frequency and phase margin are fulfilled, and furthermore, the phase derivative w. r. t. the frequency is zero, i.e., the phase Bode plot is flat at the given gain crossover frequency. Consequently, the closed-loop system is robust to gain variations. The FOC design method proposed in the paper is practical and simple to apply. Simulation and experimental results show that the closed-loop system can achieve favorable dynamic performance and robustness.

Journal ArticleDOI
TL;DR: Experimental results show that this framework yields a robust efficient on-board vehicle recognition and tracking system with high precision, high recall, and good localization.
Abstract: This paper introduces a general active-learning framework for robust on-road vehicle recognition and tracking. This framework takes a novel active-learning approach to building vehicle-recognition and tracking systems. A passively trained recognition system is built using conventional supervised learning. Using the query and archiving interface for active learning (QUAIL), the passively trained vehicle-recognition system is evaluated on an independent real-world data set, and informative samples are queried and archived to perform selective sampling. A second round of learning is then performed to build an active-learning-based vehicle recognizer. Particle filter tracking is integrated to build a complete multiple-vehicle tracking system. The active-learning-based vehicle-recognition and tracking (ALVeRT) system has been thoroughly evaluated on static images and roadway video data captured in a variety of traffic, illumination, and weather conditions. Experimental results show that this framework yields a robust efficient on-board vehicle recognition and tracking system with high precision, high recall, and good localization.


Journal ArticleDOI
TL;DR: In this paper, an evolutionary-based approach to solve the optimal power flow (OPF) problem is presented, which employs differential evolution algorithm for optimal settings of OPF problem control variables.

Proceedings ArticleDOI
01 Sep 2010
TL;DR: The model proposed does not suffer from the limitation of diminishing step size in gradient searching and allows fast asymptotic convergence, and shows robustness to additive noise, which is a main curse for algorithms based on convex mixing or consensus.
Abstract: In this paper, we propose a novel computation model for solving the distributed optimization problem where the objective function is formed by the sum of convex functions available to individual agent. Our approach differentiates from the existing approach by local convex mixing and gradient searching in that we force the states of the model to the global optimal point by controlling the subgradient of the global optimal function. In this way, the model we proposed does not suffer from the limitation of diminishing step size in gradient searching and allows fast asymptotic convergence. The model also shows robustness to additive noise, which is a main curse for algorithms based on convex mixing or consensus.

Journal ArticleDOI
TL;DR: The test results presented in this paper show the validity of the proposed low-cost sensorless-control algorithm and underline the high dynamic performances of the sensor less-control system also with a reduced equipment.
Abstract: In this paper, a low-time-consuming and low-cost sensorless-control algorithm for high-dynamic performance permanent-magnet synchronous motors, both surface and internal permanent-magnet mounted for position and speed estimation, is introduced, discussed, and experimentally validated. This control algorithm is based on the estimation of rotor speed and angular position starting from the back electromotive force space-vector determination without voltage sensors by using the reference voltages given by the current controllers instead of the actual ones. This choice obviously introduces some errors that must be vanished by means of a compensating function. The novelties of the proposed estimation algorithm are the position-estimation equation and the process of compensation of the inverter phase lag that also suggests the final mathematical form of the estimation. The mathematical structure of the estimation guarantees a high degree of robustness against parameter variation as shown by the sensitivity analysis reported in this paper. Experimental verifications of the proposed sensorless-control system have been made with the aid of a flexible test bench for brushless motor electrical drives. The test results presented in this paper show the validity of the proposed low-cost sensorless-control algorithm and, above all, underline the high dynamic performances of the sensorless-control system also with a reduced equipment.

Journal ArticleDOI
TL;DR: An open source implementation of flux variability analysis called fastFVA is presented, which makes large-scale flux variabilityAnalysis feasible and tractable allowing more complex biological questions regarding network flexibility and robustness to be addressed.
Abstract: Background Flux variability analysis is often used to determine robustness of metabolic models in various simulation conditions. However, its use has been somehow limited by the long computation time compared to other constraint-based modeling methods.

Journal ArticleDOI
TL;DR: A neural-network-based adaptive approach is proposed for the leader-following control of multiagent systems that takes uncertainty in the agent's dynamics into account; the leader's state could be time-varying; and the proposed algorithm for each following agent is only dependent on the information of its neighbor agents.
Abstract: A neural-network-based adaptive approach is proposed for the leader-following control of multiagent systems. The neural network is used to approximate the agent's uncertain dynamics, and the approximation error and external disturbances are counteracted by employing the robust signal. When there is no control input constraint, it can be proved that all the following agents can track the leader's time-varying state with the tracking error as small as desired. Compared with the related work in the literature, the uncertainty in the agent's dynamics is taken into account; the leader's state could be time-varying; and the proposed algorithm for each following agent is only dependent on the information of its neighbor agents. Finally, the satisfactory performance of the proposed method is illustrated by simulation examples.

Journal ArticleDOI
01 Jan 2010
TL;DR: It is proven that all the signals of the resulting closed-loop system are uniformly bounded and that the tracking errors converge to a small neighborhood around zero.
Abstract: A robust adaptive fuzzy control approach is developed for a class of multi-input-multi-output (MIMO) nonlinear systems with modeling uncertainties and external disturbances by using both the approximation property of the fuzzy logic systems and the backstepping technique. The MIMO systems are composed of interconnected subsystems in the strict-feedback form. The main characteristics of the developed approach are that the online computation burden is alleviated and the robustness to dynamic uncertainties and external disturbances is improved. It is proven that all the signals of the resulting closed-loop system are uniformly bounded and that the tracking errors converge to a small neighborhood around zero. Two simulation experiments are presented to demonstrate the feasibility of the approach developed in this paper.

Journal ArticleDOI
TL;DR: In this paper, the authors define the notion of risk measurement procedure, which includes both of these steps, and introduce a rigorous framework for studying the robustness of risk measurements and their sensitivity to changes in the data set.
Abstract: Measuring the risk of a financial portfolio involves two steps: estimating the loss distribution of the portfolio from available observations and computing a ‘risk measure’ that summarizes the risk of the portfolio. We define the notion of ‘risk measurement procedure’, which includes both of these steps, and introduce a rigorous framework for studying the robustness of risk measurement procedures and their sensitivity to changes in the data set. Our results point to a conflict between the subadditivity and robustness of risk measurement procedures and show that the same risk measure may exhibit quite different sensitivities depending on the estimation procedure used. Our results illustrate, in particular, that using recently proposed risk measures such as CVaR/expected shortfall leads to a less robust risk measurement procedure than historical Value-at-Risk. We also propose alternative risk measurement procedures that possess the robustness property.

Journal ArticleDOI
TL;DR: By using finite-time stability theorem, inequality techniques, the properties of Weiner process and adding suitable controllers, sufficient conditions are obtained to ensure finite- time stochastic synchronization for the complex networks.

Journal ArticleDOI
01 Aug 2010
TL;DR: Traditional L2-norm-based least squares criterion is sensitive to outliers, while the newly proposed L1-norm 2DPCA is robust.
Abstract: In this paper, we first present a simple but effective L1-norm-based two-dimensional principal component analysis (2DPCA). Traditional L2-norm-based least squares criterion is sensitive to outliers, while the newly proposed L1-norm 2DPCA is robust. Experimental results demonstrate its advantages.

Journal ArticleDOI
TL;DR: In this paper, the robustness of positive-position feedback control of flexible structures with colocated force actuators and position sensors is investigated using the theory of negative-imaginary systems, which can be extended to nonlinear systems through the notion of counterclockwise input-output dynamics.
Abstract: This paper investigates the robustness of positive-position feedback control of flexible structures with colocated force actuators and position sensors. In particular, the theory of negative-imaginary systems is used to reveal the robustness properties of multi-input, multi-output (MIMO) positive-position feedback controllers and related types of controllers for flexible structures. The negative-imaginary property of linear systems can be extended to nonlinear systems through the notion of counterclockwise input-output dynamics.

Journal ArticleDOI
TL;DR: It is demonstrated how an ideal capacity-disruption range can be objectively determined for a particular network and introduced a scalable system-wide performance measure, called the Network Trip Robustness (NTR), that can be used to directly compare networks of different sizes, topologies, and connectivity levels.
Abstract: A wide range of relatively short-term disruptive events such as partial flooding, visibility reductions, traction hazards due to weather, and pavement deterioration occur on transportation networks on a daily basis. Despite being relatively minor when compared to catastrophes, these events still have profound impacts on traffic flow. To date there has been very little distinction drawn between different types of network-disruption studies and how the methodological approaches used in those studies differ depending on the specific research objectives and on the disruption scenarios being modeled. In this paper, we advance a methodological approach that employs different link-based capacity-disruption values for identifying and ranking the most critical links and quantifying network robustness in a transportation network. We demonstrate how an ideal capacity-disruption range can be objectively determined for a particular network and introduce a scalable system-wide performance measure, called the Network Trip Robustness (NTR) that can be used to directly compare networks of different sizes, topologies, and connectivity levels. Our approach yields results that are independent of the degree of connectivity and can be used to evaluate robustness on networks with isolating links. We show that system-wide travel-times and the rank-ordering of the most critical links in a network can vary dramatically based on both the capacity-disruption level and on the overall connectivity of the network. We further show that the relationships between network robustness, the capacity-disruption level used for modeling, and network connectivity are non-linear and not necessarily intuitive. We discuss our findings with respect to Braess' Paradox.

Book ChapterDOI
05 Sep 2010
TL;DR: This work proposes a novel approach to providing robustness to both occlusions and viewpoint changes that yields significant improvements over existing techniques in action recognition.
Abstract: Most state-of-the-art approaches to action recognition rely on global representations either by concatenating local information in a long descriptor vector or by computing a single location independent histogram. This limits their performance in presence of occlusions and when running on multiple viewpoints. We propose a novel approach to providing robustness to both occlusions and viewpoint changes that yields significant improvements over existing techniques. At its heart is a local partitioning and hierarchical classification of the 3D Histogram of Oriented Gradients (HOG) descriptor to represent sequences of images that have been concatenated into a data volume. We achieve robustness to occlusions and viewpoint changes by combining training data from all viewpoints to train classifiers that estimate action labels independently over sets of HOG blocks. A top level classifier combines these local labels into a global action class decision.