Light field imaging has emerged as a technology allowing to capture richer visual information from our world. As opposed traditional photography, which captures 2D projection of the light in scene integrating angular domain, fields collect radiance rays all directions, demultiplexing lost conventional photography. On one hand, this higher dimensional representation data offers powerful capabilities for understanding, and substantially improves performance computer vision problems such depth...

In this paper, a decentralized fault diagnosis approach of complex processes is proposed based on multiblock kernel partial least squares (MBKPLS). To solve the problem posed by nonlinear characteristics, (KPLS) approaches have been proposed. MBKPLS algorithm first and applied to monitor large-scale processes. The advantages are: 1) can capture more useful information between within blocks compared (PLS); 2) gives interpretation MBPLS; 3) Fault becomes possible if number sub-blocks equal...

Conventional evolutionary algorithms (EAs) are not well suited for solving expensive optimization problems due to the fact that they often require a large number of fitness evaluations obtain acceptable solutions. To alleviate difficulty, this paper presents multitasking framework computationally problems. In framework, knowledge is transferred from cheap help solution problem on basis recently proposed multifactorial EA (MFEA), leading faster convergence problem. However, existing MFEAs do...

To perform power augmentation tasks of a robotic exoskeleton, this paper utilizes fuzzy approximation and designed disturbance observers to compensate for the torques caused by unknown input saturation, errors, viscous friction, gravity, payloads. The proposed adaptive control with updated parameters' mechanism additional torque inputs using are exerted into exoskeleton via feedforward loops counteract disturbances. Through such an approach, system does not need any requirement built-in...

In this paper, we take advantage of the clear texture structure epipolar plane image (EPI) in light field data and model problem reconstruction from a sparse set views as CNN-based angular detail restoration on EPI. We indicate that one main challenges sparsely sampled is information asymmetry between spatial domain, where portion domain damaged by undersampling. To balance information, high frequency components an EPI removed using blur, before feeding to network. Finally, non-blind deblur...

Gaussian processes (GPs) are the most popular model used in surrogate-assisted evolutionary optimization of computationally expensive problems, mainly because GPs able to measure uncertainty estimated fitness values, based on which certain infill sampling criteria can be guide search and update surrogate model. However, computation time for constructing may become excessively long when number training samples increases, makes it inappropriate use them as surrogates optimization. To address...

Event-triggered control is an effective strategy, capable of reducing the amount communications and retaining a satisfactory closed-loop performance. In this paper, we aim at proposing event-triggered sampling mechanism studying observer-based output feedback for switched linear neutral systems with mixed time-varying delays. Different from conventional strategies, our proposed one transmits not only state but also switching information to controller, which advantageous in applications where...

Based on the analysis of characteristics and operation status process industry, as well development global intelligent manufacturing a new mode for namely, deep integration industrial artificial intelligence Industrial Internet with is proposed. This paper analyzes existing three-tier structure which consists enterprise resource planning, execution system, control examines decision-making, control, management adopted by enterprises. this analysis, it then describes meaning an framework...

Gaussian processes (GPs) are widely used in surrogate-assisted evolutionary optimization of expensive problems mainly due to the ability provide a confidence level their outputs, making it possible adopt principled surrogate management methods, such as acquisition function Bayesian optimization. Unfortunately, GPs become less practical for high-dimensional multiobjective and many-objective computational complexity is cubic number training samples. In this article, we propose computationally...

This paper is concerned with the trajectory tracking control problem for wheeled mobile robots (WMRs) subject to actuator faults. The challenge lies in partial loss of effectiveness actuated wheels which results strong controllability WMR, rendering classical fault-tolerant methods infeasible. To overcome this obstacle, a novel mixed-gain adaption technique put forward and skillfully combined robust prescribed performance method. resulting achieves WMR predefined settling time accuracy,...

<para xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> This paper develops a guaranteed cost networked control (GCNC) method for Takagi–Sugeno (T–S) fuzzy systems with time delays. The state feedback controller is designed via the system (NCS) theory. stability of overall using GCNC also established. Network-induced delay in network transmission and packet dropout are analyzed. Some deductions extended to uncertain systems. Simulation results show...

This paper investigates the problems of passivity analysis and passification for network-based linear control systems. A new sampled-data model is first formulated based on updating instants ZOH (zeroth order hold), where physical plant controller are, respectively, in continuous time discrete time. In this model, network-induced delays, data packet dropouts, signal measurement quantization have been taken into account. The quantizer assumed to be logarithmic, delays are both a lower bound...

This paper develops an off-policy reinforcement learning (RL) algorithm to solve optimal synchronization of multiagent systems. is accomplished by using the framework graphical games. In contrast traditional control protocols, which require complete knowledge agent dynamics, proposed RL a model-free approach, in that it solves problem without knowing any dynamics. A prescribed policy, called behavior applied each generate and collect data for learning. An Bellman equation derived learn value...

In this paper, global adaptive finite-time stabilization is investigated by logic-based switching control for a class of uncertain nonlinear systems with the powers positive odd rational numbers. Parametric uncertainties entering state equations nonlinearly can be fast time-varying or jumping at unknown time instants, and coefficient appearing in channel unknown. The bounds parametric are not required to know priori. Our proposed controller switching-type one, which two parameters tuned...

This paper studies the operational optimal control problem for industrial flotation process, a key component in mineral processing concentrator line. A new model-free data-driven method is developed here real-time solution of this problem. novel formulation given selection process inputs that guarantees tracking indices while maintaining within specified bounds. Proper prescribed indices, namely concentrate grade and tail grade, essential proper economic operation process. The difficulty...

This paper investigates the setpoints compensation for a class of complex industrial processes. Plants at device layer are controlled by local regulation controllers, and multirate output feedback control approach is proposed such that plants can reach dynamically changed given economic objective also be tracked via certain performance index (EPI). First, sampled-data multivariable proportional integral (PI) controller designed to regulate plants. Second, outputs inputs sampled operation...

In this paper, a novel convolutional neural network (CNN)-based framework is developed for light field reconstruction from sparse set of views. We indicate that the can be efficiently modeled as angular restoration on an epipolar plane image (EPI). The main problem in direct EPI involves information asymmetry between spatial and dimensions, where detailed portion dimensions damaged by undersampling. Directly upsampling or super-resolving causes ghosting effects. To suppress these effects, we...

In this paper, a novel off-policy interleaved Q-learning algorithm is presented for solving optimal control problem of affine nonlinear discrete-time (DT) systems, using only the measured data along system trajectories. Affine feature unknown dynamics, and learning approach pose tremendous challenges on approximating controllers. To end, on-policy method DT systems reviewed first, its convergence rigorously proven. The bias solution to Q-function-based Bellman equation caused by adding...

Principal component analysis (PCA) and independent (ICA) have been widely used for process monitoring in industry. Since the operation data of blast furnace (BF) ironmaking contain both non-Gaussian distribution Gaussian data, above single PCA or ICA method hardly describes information BF completely, which makes diagnosis abnormal working-conditions only with a prone to false positives negatives. In this article, novel integrated PCA-ICA is proposed diagnosing conditions by comprehensively...

Optimal operation of an industrial blast furnace (BF) ironmaking process largely depends on a reliable measurement molten iron quality (MIQ) indices, which are not feasible using the conventional sensors. This paper proposes novel data-driven robust modeling method for online estimation and control MIQ indices. First, nonlinear autoregressive exogenous (NARX) model is constructed indices to completely capture dynamics BF process. Then, considering that standard least-squares support vector...