Internal State Of System Research Articles

Dealing with Complexity through Advanced Control Techniques

The control of complex dynamic systems, both in their behavior and in their mission, goes through the implementation of multi-loop control architectures based on information about the system internal state and from the environment, as well as on the mission plan state. This results in systems that are becoming increasingly autonomous, for which requirements in terms of safety and reliability, as well as expected performance, are increasingly high. Research works developed at ONERA in the field of control for autonomous systems cover all levels of the control architectures, which are basically structured with respect to temporal aspects, as well as the level of abstraction that they entail for the system dynamic. We will consider them in this paper by increasing level. We will discuss the advances achieved recently in the robust control techniques of uncertain dynamic systems generally implemented at the lower control level and we will discuss their extensions to consider input and output constraints, as well as the hybrid nature of most of the systems considered. To design task level control primitives that take place just above the previous control loops, we will introduce sensor-based robust and non-linear control techniques. These are based on information on the environment extracted from exteroceptive sensors, to adapt system behavior to uncertainties and perturbations. Multi-sensor and/or multi-objective controls will be discussed in this particular context. We will also present several recent results in the field of trajectory tracking based on visual navigation techniques in complex environments, which combine objectives and constraints within the same control architecture. We will discuss how model predictive control (MPC) techniques and advanced optimization techniques can be used for solving the resulting control problems. In addition, we will discuss several ongoing developments of these methods by exploiting distributed model predictive control techniques (DMPC) and predictive control of hybrid systems. Finally, integration with the control architectures at the upper level of reactive, predictive and distributed planning capabilities will be proposed to accommodate time constraints and uncertainties in decision.

Characterization of complex states for friction‐excited systems

AbstractFor many engineering structures the dynamical response during operation is rarely stationary or regular. In fact, many studies on measured data reveal the irregular nature of dynamical system responses, for example in the field of research on brake squeal. Therefore, linear measures, such as amplitude and frequency spectra, cannot characterize the core dynamics contained in an irregular time series. Additionally, linear methods for damping extraction from measured signals tend to lose significance when it comes to irregular responses of self‐ excited systems. This work aims at highlighting the need for quantifiers from nonlinear time series analysis to adequately characterize real‐world vibration data. In a first step a minimal model is studied to relate nonlinear time series quantifiers to internal system states and dissipation considerations. The goal of this research is to develop an understanding for the information contained in time series data concerning dissipation mechanisms of complex friction‐excited systems. (© 2017 Wiley‐VCH Verlag GmbH & Co. KGaA, Weinheim)

Discrimination of singleton and periodic attractors in Boolean networks

Determining the minimum number of sensor nodes to observe the internal state of the whole system is important in analysis of complex networks. However, existing studies suggest that a large number of sensor nodes are needed to know the whole internal state. In this paper, we focus on identification of a small set of sensor nodes to discriminate statically and periodically steady states using the Boolean network model where steady states are often considered to correspond to cell types. In other words, we seek a minimum set of nodes to discriminate singleton and periodic attractors. We prove that one node is not necessarily enough but two nodes are always enough to discriminate two periodic attractors by using the Chinese remainder theorem. Based on this, we present an algorithm to determine the minimum number of nodes to discriminate all given attractors. We also present a much more efficient algorithm to discriminate singleton attractors. The results of computational experiments suggest that attractors in realistic Boolean networks can be discriminated by observing the states of only a small number of nodes.

Event-Triggered Distributed Approximate Optimal State and Output Control of Affine Nonlinear Interconnected Systems.

This paper presents an approximate optimal distributed control scheme for a known interconnected system composed of input affine nonlinear subsystems using event-triggered state and output feedback via a novel hybrid learning scheme. First, the cost function for the overall system is redefined as the sum of cost functions of individual subsystems. A distributed optimal control policy for the interconnected system is developed using the optimal value function of each subsystem. To generate the optimal control policy, forward-in-time, neural networks are employed to reconstruct the unknown optimal value function at each subsystem online. In order to retain the advantages of event-triggered feedback for an adaptive optimal controller, a novel hybrid learning scheme is proposed to reduce the convergence time for the learning algorithm. The development is based on the observation that, in the event-triggered feedback, the sampling instants are dynamic and results in variable interevent time. To relax the requirement of entire state measurements, an extended nonlinear observer is designed at each subsystem to recover the system internal states from the measurable feedback. Using a Lyapunov-based analysis, it is demonstrated that the system states and the observer errors remain locally uniformly ultimately bounded and the control policy converges to a neighborhood of the optimal policy. Simulation results are presented to demonstrate the performance of the developed controller.

A Comparison of Predictive Parameter Estimation using Kalman Filter and Analysis of Variance

The design of a controller significantly improves if internal states of a dynamic control system are predicted. This paper compares the prediction of system states using Kalman filter and a novel approach analysis of variance (ANOVA). Kalman filter has been successfully applied in several applications. A significant advantage of Kalman filter is its ability to use system output to predict the future states. It has been observed that Kalman filter based predictive controller design outperforms many other approaches. An important drawback of such controllers is however that their performances deteriorate in situations where the system states have no correlation with the output. This paper takes a hypothetical model of a helicopter and builds system model using the state-space diagram. The design is implemented using SIMULINK. It has been observed that in situations where the states are dependent on system output, the ANOVA based state prediction gives comparable results with that of Kalman filter based parameter estimation. The ANOVA based parameter prediction, however outperforms Kalman filter based parameter prediction in situations where the output does not directly contribute in a particular state. The research was based on empirical results. Rigorous testing was performed on four internal states to prove that ANOVA based predictive parameter estimation technique outperforms Kalman based parameter estimation in situations where the system internal states is not directly linked with the output.

Kalman Filtering Solution Converges on a Personal Computer

Instantaneous observability is used to watch a system output with very fast signals as well as it is a system property that enables to estimate system internal states. This property depends on the pair of discrete matrices [Formula: see text] and it considers that the system state equations are known. The problem is that the system states are inside and they are not always accessible directly. A process, which is a time-varying running program in four parts composes the system under investigation here. It is shown it is possible to apply Kalman filtering on a digital personal computer’s system with particularly the four parts like the ones under investigation. A computing process is performed during a period of time called latency. The calculation of latency considers it as a random variable with Gaussian distribution. The potential application of the results attained is the forecasting of data traffic-jam on a digital personal computer, which has very fast signals inside. In a broader perspective, this method to calculate latency can be applied on other digital personal computer processes such as processes on random access memory. It is also possible to apply this method on local area networks and mainframes.

A hybrid framework combining data-driven and model-based methods for system remaining useful life prediction

Remaining useful life prediction is one of the key requirements in prognostics and health management. While a system or component exhibits degradation during its life cycle, there are various methods to predict its future performance and assess the time frame until it does no longer perform its desired functionality. The proposed data-driven and model-based hybrid/fusion prognostics framework interfaces a classical Bayesian model-based prognostics approach, namely particle filter, with two data-driven methods in purpose of improving the prediction accuracy. The first data-driven method establishes the measurement model (inferring the measurements from the internal system state) to account for situations where the internal system state is not accessible through direct measurements. The second data-driven method extrapolates the measurements beyond the range of actually available measurements to feed them back to the model-based method which further updates the particles and their weights during the long-term prediction phase. By leveraging the strengths of the data-driven and model-based methods, the proposed fusion prognostics framework can bridge the gap between data-driven prognostics and model-based prognostics when both abundant historical data and knowledge of the physical degradation process are available. The proposed framework was successfully applied on lithium-ion battery remaining useful life prediction and achieved a significantly better accuracy compared to the classical particle filter approach.

An Event-Triggered ADP Control Approach for Continuous-Time System With Unknown Internal States.

This paper proposes a novel event-triggered adaptive dynamic programming (ADP) control method for nonlinear continuous-time system with unknown internal states. Comparing with the traditional ADP design with a fixed sample period, the event-triggered method samples the state and updates the controller only when it is necessary. Therefore, the computation cost and transmission load are reduced. Usually, the event-triggered method is based on the system entire state which is either infeasible or very difficult to obtain in practice applications. This paper integrates a neural-network-based observer to recover the system internal states from the measurable feedback. Both the proposed observer and the controller are aperiodically updated according to the designed triggering condition. Neural network techniques are applied to estimate the performance index and help calculate the control action. The stability analysis of the proposed method is also demonstrated by Lyapunov construct for both the continuous and jump dynamics. The simulation results verify the theoretical analysis and justify the efficiency of the proposed method.

Disturbance Observer Design for Utilizing of Time-delayed Vision Measurements in High Dynamical Systems

The utilization of vision information for control in intelligent technical applications with high dynamics has been a central issue. The main obstacles are the low sampling rate of the cameras (about 60Hz of a commercial camera) compared to the required sampling rate of the high dynamic applications (over 1kHz) and the time-delayed measurements due to the image processing. A model-based Event-Triggered Observer (ETO) [12] has been proposed for linear systems to observe and predict the undelayed continuous state variables of the system from the sampled and delayed measurements where the sampling rate and the delay- time can be constant or variable. However, the disturbances like the sensor noise and the perturbations of the environment were not handled within [12]. As is well known, the disturbances decline the performance of the model-based observer. To keep the performance of the state-estimation and prediction, an effective solution is to adopt into account the unknown dynamic effects due to the disturbances by integrating them into the state-estimation. A survey [13] has shown different methods of dealing with the disturbance estimation to improve the accuracy of the estimation of the system internal states. Unlike other mentioned methods in [13], the so-called Unknown Input Observer (UIO), shows significant advantages with the ability to estimate the unknown input disturbances and the system internal states simultaneously by augmenting the state space of a classical state observer like Luenberger-Observer with a disturbance model [14,15]. In this paper, an Extended Event-Triggered Observer (EETO) is proposed, which based on the proposed ETO in [12] and is extended with an UIO. The new proposed EETO can estimate undelayed continuous state variables from the time-delayed measurements under the disturbances. An example system with a second-order delay behavior is simulated, whose output is sampled and delayed. The simulation results show the performance of the proposed Extended Event-Triggered Observer.

Study on the Environmental Cognition of a Self-evolving Conscious System

This study proposes a conscious model for a robot that achieves self-evolution by evaluating whether the internal state of the conscious system is represented as pleasant or unpleasant. For the definition of pleasant and unpleasant, we adopted the “smoothness of the flow of information” in the conscious system. Thus, our proposed conscious robot is capable of self-evolving tailored to the environment in which the robot itself is placed. In other words, it is possible to identify whether or not any external information has been input, and to identify the information with time series memory.

Specification and Synthesis of Reactive Protocols for Aircraft Electric Power Distribution

The increasing complexity of electric power systems leads to challenges in integration and verification. We consider the problem of designing a control protocol for an aircraft electric power system that meets a set of requirements describing the correct behaviors of the system and reacts dynamically to changes in internal system states. We formalize the requirements by translating them into a temporal logic specification language and apply game-based, temporal logic formal methods to automatically synthesize a controller protocol that satisfies these overall properties and requirements. Through a case study, we perform a design exploration to show the benefits and tradeoffs between centralized and distributed control architectures.

Adaptive relevance feedback for large-scale image retrieval

Content-based image retrieval aims at substituting traditional indexing based on manual annotation by using automatically-extracted visual indexing features. Novel techniques are needed however to efficiently deal with the semantic gap (i.e. the partial match between the low-level features and the visual content). Here, we investigate a query-free retrieval approach first proposed by Ferecatu and Geman. This approach relies solely on an iterative relevance feedback mechanism that drives a heuristic sampling of the collection, and aims to take explicitly into account the semantic gap. Our contributions are related to three complementary aspects. First, we formalize a large-scale approach based on a hierarchical tree-like organization of the images computed off-line. Second, we propose a versatile modulation of the exploration/exploitation trade-off based on the consistency of the system internal states between successive iterations. Third, we elaborate a long-term optimization of the similarity metric based on the user searching session logs accumulated off-line. We implemented a web-application that integrates all our contributions, and distribute it under the AGPL Version 3 free software license. We organized user-based evaluation campaigns using ImageNet dataset, and show empirically that our contributions significantly improve the retrieval performance of the original framework, that they are complementary to each other, and that their overall integration is consistently beneficial.

Realizing actual feedback control of complex network

Abstract In this paper, we present the concept of feedbackability and how to identify the Minimum Feedbackability Set of an arbitrary complex directed network. Furthermore, we design an estimator and a feedback controller accessing one MFS to realize actual feedback control, i.e. control the system to our desired state according to the estimated system internal state from the output of estimator. Last but not least, we perform numerical simulations of a small linear time-invariant dynamics network and a real simple food network to verify the theoretical results. The framework presented here could make an arbitrary complex directed network realize actual feedback control and deepen our understanding of complex systems.

Software rejuvenation in cluster computing systems with dependency between nodes

Software rejuvenation is a preventive and proactive fault management technique that is particularly useful for counteracting the phenomenon of software aging, aimed at cleaning up the system internal state to prevent the occurrence of future failure. The increasing interest in combing software rejuvenation with cluster systems has given rise to a prolific research activity in recent years. However, so far there have been few reports on the dependency between nodes in cluster systems when software rejuvenation is applied. This paper investigates the software rejuvenation policy for cluster computing systems with dependency between nodes, and reconstructs an stochastic reward net model of the software rejuvenation in such cluster systems. Simulation experiments and results reveal that the software rejuvenation strategy can decrease the failure rate and increase the availability of the cluster system. It also shows that the dependency between nodes affects software rejuvenation policy. Based on the theoretic analysis of the software rejuvenation model, a prototype is implemented on the Smart Platform cluster computing system. Performance measurement is carried out on this prototype, and experimental results reveal that software rejuvenation can effectively prevent systems from entering into disabled states, and thereby improving the ability of software fault-tolerance and the availability of cluster computing systems.

A Comparison Between the Human Sense of Smell and Neural Activity in the Olfactory Bulb of Rats

Generally, odor qualities are evaluated via sensory tests in which predefined criteria are assessed by panelists and stochastically analyzed to reduce human inconsistencies. Because this method requires multiple, well-trained human subjects, a more convenient approach is required to enable predictions of odor qualities. In this article, we propose an approach involving linking internal states of the olfactory system with perceptual characteristics. In the study, the glomerular responses of rats were taken to represent internal olfactory system states. Similarities between the glomerular responses of rats were quantified by correlations between glomerular activity patterns, overlap rate of strongly activated part across glomerular activity patterns, and the similarity between histograms of the strength of activity. These indices were then compared with perceptual similarities measured from human subjects in sensory tests. The results of experiments involving 22 odorants showed medium strength correlations between each index and perceptual similarity. In addition, when the 3 indices were combined using their Euclidean distance, we observed middle to high correlations (r = 0.65-0.79) to human perceptual similarity. We also report the results of our use of a machine learning technique to classify the odorants into a similar and dissimilar category. Although the correct rate of classification varied from 33.3% to 92.9%, these results support the feasibility of linking the glomerular responses of rats to human perception.

Nonlinear dynamics of complex hysteretic systems: Oscillator in a magnetic field

Complex hysteresis is a well-known phenomenon in many branches of science. The most prominent examples come from materials with a complex microscopic structure such as magnetic materials, shape-memory alloys, or, porous materials. Their hysteretic behavior is characterized by the existence of multiple internal system states for a given external parameter and by a non-local memory. The input-output behavior of such systems is well studied and in a standard phenomenological approach described by the so-called Preisach operator. What is not well understood, are situations, where such a hysteretic system is dynamically coupled to its environment. Since the hysteretic sub-system provides a complicated form of nonlinearity, one expects non-trivial, possibly chaotic behavior of the combined dynamical system. We study such a combined dynamical system with hysteretic nonlinearity. In this original contribution a simple differential-operator equation with hysteretic damping, which describes a magnetic pendulum is considered. We find, for instance, a fractal dependence of the asymptotic behavior as function of the starting values. The sensitivity of the system to perturbations is investigated by several methods, such as the 0–1 test for chaos and sub-Lyapunov exponents. The power spectral density is also calculated and compared with analytical results for simple input-output scenarios.

Workload-Based Software Rejuvenation in Cloud Systems

Cloud computing is a promising paradigm able to rationalize the use of hardware resources by means of virtualization. Virtualization allows to instantiate one or more virtual machines (VMs) on top of a single physical machine managed by a virtual machine monitor (VMM). Similarly to any other software, a VMM experiences aging and failures. Software rejuvenation is a proactive fault management technique that involves terminating an application, cleaning up the system internal state, and restarting it to prevent the occurrence of future failures. In this work, we propose a technique to model and evaluate the VMM aging process and to investigate the optimal rejuvenation policy that maximizes the VMM availability under variable workload conditions. Starting from dynamic reliability theory and adopting symbolic algebraic techniques, we investigate and compare existing time-based VMM rejuvenation policies. We also propose a time-based policy that adapts the rejuvenation timer to the VMM workload condition improving the system availability. The effectiveness of the proposed modeling technique is demonstrated through a numerical example based on a case study taken from the literature.

Observability of complex systems

A quantitative description of a complex system is inherently limited by our ability to estimate the system's internal state from experimentally accessible outputs. Although the simultaneous measurement of all internal variables, like all metabolite concentrations in a cell, offers a complete description of a system's state, in practice experimental access is limited to only a subset of variables, or sensors. A system is called observable if we can reconstruct the system's complete internal state from its outputs. Here, we adopt a graphical approach derived from the dynamical laws that govern a system to determine the sensors that are necessary to reconstruct the full internal state of a complex system. We apply this approach to biochemical reaction systems, finding that the identified sensors are not only necessary but also sufficient for observability. The developed approach can also identify the optimal sensors for target or partial observability, helping us reconstruct selected state variables from appropriately chosen outputs, a prerequisite for optimal biomarker design. Given the fundamental role observability plays in complex systems, these results offer avenues to systematically explore the dynamics of a wide range of natural, technological and socioeconomic systems.

Quantum systems in a stationary environment out of thermal equilibrium

We discuss how the thermalization of an elementary quantum system is modified when the system is placed in an environment out of thermal equilibrium. To this aim we provide a detailed investigation of the dynamics of an atomic system placed close to a body of arbitrary geometry and dielectric permittivity, whose temperature $T_M$ is different from that of the surrounding walls $T_W$. A suitable master equation for the general case of an $N$-level atom is first derived and then specialized to the cases of a two- and three-level atom. Transition rates and steady states are explicitly expressed as a function of the scattering matrices of the body and become both qualitatively and quantitatively different from the case of radiation at thermal equilibrium. Out of equilibrium, the system steady state depends on the system-body distance, on the geometry of the body and on the interplay of all such parameters with the body optical resonances. While a two-level atom tends toward a thermal state, this is not the case already in the presence of three atomic levels. This peculiar behavior can be exploited, for example, to invert the populations ordering and to provide an efficient cooling mechanism for the internal state of the quantum system. We finally provide numerical studies and asymptotic expressions when the body is a slab of finite thickness. Our predictions can be relevant for a wide class of experimental configurations out of thermal equilibrium involving different physical realizations of two or three-level systems.

Software Aging Analysis of Web Server Using Neural Networks

Software aging is a phenomenon that refers to progressive performance degradation or transient failures or even crashes in long running software systems such as web servers. It mainly occurs due to the deterioration of operating system resource, fragmentation and numerical error accumulation. A primitive method to fight against software aging is software rejuvenation. Software rejuvenation is a proactive fault management technique aimed at cleaning up the system internal state to prevent the occurrence of more severe crash failures in the future. It involves occasionally stopping the running software, cleaning its internal state and restarting it. An optimized schedule for performing the software rejuvenation has to be derived in advance because a long running application could not be put down now and then as it may lead to waste of cost. This paper proposes a method to derive an accurate and optimized schedule for rejuvenation of a web server (Apache) by using Radial Basis Function (RBF) based Feed Forward Neural Network, a variant of Artificial Neural Networks (ANN). Aging indicators are obtained through experimental setup involving Apache web server and clients, which acts as input to the neural network model. This method is better than existing ones because usage of RBF leads to better accuracy and speed in convergence.