In this paper, an adaptive fuzzy optimal control design is addressed for a class of unknown nonlinear discrete-time systems. The controlled systems are in strict-feedback frame and contain functions nonsymmetric dead-zone. For systems, the objective to controller, which not only guarantees stability but achieves performance as well. This immediately brings about difficulties controller design. To end, logic employed approximate Based on utility critic designs, by applying backsteppping...

The application of multiobjective evolutionary algorithms to many-objective optimization problems often faces challenges in terms diversity and convergence. On the one hand, with a limited population size, it is difficult for an algorithm cover different parts whole Pareto front (PF) large objective space. tends concentrate only on areas. other as number objectives increases, solutions easily have poor values some objectives, which can be regarded bottleneck that restrict solutions'...

In this article, a novel reinforcement learning-based optimal tracking control (RLOTC) scheme is established for an unmanned surface vehicle (USV) in the presence of complex unknowns, including dead-zone input nonlinearities, system dynamics, and disturbances. To be specific, nonlinearities are decoupled to input-dependent sloped controls unknown biases that encapsulated into lumped unknowns within error dynamics. Neural network (NN) approximators further deployed adaptively identify...

In this paper, an adaptive fuzzy controller is constructed for a class of nonlinear discrete-time systems with unknown functions and bounded disturbances. The main characteristics the are that they take into account effect dead zone system states not required to be measurable. stability problem first time addressed in paper. Due unavailability presence zone, design becomes more difficult. To stabilize uncertain systems, logic used approximate functions, state observer designed estimate...

An unmanned surface vehicle (USV) under complicated marine environments can hardly be modeled well such that model-based optimal control approaches become infeasible. In this article, a self-learning-based model-free solution only using input–output signals of the USV is innovatively provided. To end, data-driven performance-prescribed reinforcement learning (DPRLC) scheme created to pursue optimality and prescribed tracking accuracy simultaneously. By devising state transformation with...

Path planning is a critical issue to ensure the safety and reliability of autonomous navigation system underwater vehicles (AUVs). Due nonlinearity constraint issues, existing algorithms perform unsatisfactorily or even cannot find feasible solution when facing large-scale problem spaces. This paper improves path AUVs in terms both model optimization algorithm. The proposed comprehensive, which aggregates length, energy consumption, collision risk into objective function incorporates...

In this paper, an effective adaptive control approach is constructed to stabilize a class of nonlinear discrete-time systems, which contain unknown functions, dead-zone input, and direction. Different from linear dead zone, the in kind zone. To overcome noncausal problem, leads scheme infeasible, systems can be transformed into $m$ -step-ahead predictor. Due appearance, predictor still contains nonaffine function. addition, it assumed that gain function input direction are unknown. These...

Security assurance in Vehicular Ad hoc Network (VANET) is a crucial and challenging task due to the open-access medium. One great threat VANETs Distributed Denial-of-Service (DDoS) attack because target of this prevent authorized nodes from accessing services. To provide high availability VANETs, scalable, reliable robust network intrusion detection system should be developed efficiently mitigate DDoS. However, big data poses serious challenges DDoS since require scalable methods capture,...

An event-triggered adaptive dynamic programming (ADP) algorithm is developed in this article to solve the tracking control problem for partially unknown constrained uncertain systems. First, an augmented system constructed, and solution of optimal transformed into regulation nominal with a discounted value function. The integral reinforcement learning employed avoid requirement drift dynamics. Second, ADP adopted its implementation, where neural network weights not only relaxes initial...

In this paper, we aim to solve the optimal tracking control problem for Henon Mapping chaotic system using direct heuristic dynamic programming (DHDP) setting with filtered error. The fuzzy logic is used approximate long-term utility function. Compared results discrete-time system, cost of controller reduced. Lyapunov analysis approach utilized prove stability system. It shown that error, adaptation law and input retain property uniformly ultimate boundedness. A simulation example given...

Although the cloud-based robotic system has provided services in various industries, its data safety is continuously threatened, and network intrusion detection (NIDS) considered as a necessary component to ensure security. In recent years, many machine learning (ML) techniques have been applied for building more intelligent NIDS. Most NIDSs based on ML method artificial intelligence are either supervised or unsupervised. However, NIDS depends much labeled data. This weakness makes it harder...

Medical image segmentation methods downsample images for feature extraction and then upsample them to restore resolution pixel-level predictions. In such schema, technique is vital in restoring information better performance. However, existing techniques leverage little from downsampling paths. The local detailed the shallower layer as boundary tissue texture crucial segmentation, especially medical segmentation. To this end, we propose a novel approach Window Attention Upsample (WAU), which...

A supply chain is a system that includes multiple entities such as suppliers, manufacturers, carriers, retailers, and customers. How to encourage the willingness cooperate share information between these has always been one of major challenges in field management. The Industrial Internet Things (IIoT) can help get real-time data key reduce costs. However, not only needs know how gather but also ensures processed leaked. Since Blockchain, main technology Bitcoin, advantages traceability...

Automation in smart manufacturing depends on many parties the supply chain and distribution networks. A manufacturer needs to track location state of materials from its suppliers products customers/distributors manage corresponding production schedule. The need for such information can be fulfilled by use Industrial Internet Things (IIoT) devices installed at each respective party. However, even though all may trust one another, they have confidential data produced IIoT that are not allowed...

Neighborhood information plays an important role in improving the performance of evolutionary computation various optimization scenarios, particularly context multimodal optimization. Several neighborhood concepts, i.e., index-based neighborhood, nearest and fuzzy have been studied engaged design niching methods. However, use these concepts requires specification some problem-related parameters, which is difficult to determine without a prior knowledge. In this paper, we introduce new...

The signal processing of industrial big data (IBD) is a challenging task, owing to the complex working scenarios and lack annotations. Defect detection, which an important subject IBD research works, has shown its effectiveness in digital inspection applications many previous studies. This article proposes novel defect detection method based on deep learning for applications. In our method, module named feature collection compression network applied merge multiscale information. Then, new...

Abstract In this article, subject to both pose and velocity constraints within narrow waters, a self‐learning‐based optimal tracking control (SLOTC) scheme is innovatively created for an unmanned surface vehicle (USV) by deploying actor‐critic reinforcement learning (RL) mechanism backstepping technique. To be specific, the barrier Lyapunov function (BLF) devised uniformly limit states predefined region pertaining smoothly feasible reference trajectory. By virtue of constrained...

The introduction of workflow in cloud computing has afforded a new and efficient way to tackle large-scale applications. As an NP-hard problem, how schedule workflows effectively economically with deadline constraints different kinds tasks resources is extraordinarily challenging. To solve this constrained paper intends develop intelligent scheduling system from the perspective users reduce expenditure workflow, subject other execution constraints. A estimation model task time designed...

Quantitative features extracted from biopsy digital pathology images can provide predictive information for neoadjuvant chemoradiotherapy (nCRT) in local advanced rectal cancer (LARC) Machine learning technologies are applied to build the digital-pathology-based signature The is an independent predictor of treatment response nCRT LARC

For the security defense in current Intelligent Transportation System (ITS), malware is often used as analysis data source, but only known attack type can be detected. A general anomaly detection framework proposed, using log source. By modeling template sequence a natural language and stacked Long Short-Term Memory (LSTM) with self-attention mechanism, effectively extract hidden pattern of sequence, well express dependencies inside sequence. The experimental results show that overall...