Intelligent Deep Learning Enabled Human Activity Recognition for Improved Medical Services

Human Activity Recognition(HAR)has been made simple in recent years,thanks to recent advancements made in Artificial Intelligence(AI)techni-ques.These techniques are applied in several areas like security,surveillance,healthcare,human-robot interaction,and entertainment.Since wearable sensor-based HAR system includes in-built sensors,human activities can be categorized based on sensor values.Further,it can also be employed in other applications such as gait diagnosis,observation of children/adult’s cognitive nature,stroke-patient hospital direction,Epilepsy and Parkinson’s disease examination,etc.Recently-developed Artificial Intelligence(AI)techniques,especially Deep Learning(DL)models can be deployed to accomplish effective outcomes on HAR process.With this motivation,the current research paper focuses on designing Intelligent Hyperparameter Tuned Deep Learning-based HAR(IHPTDL-HAR)technique in healthcare environment.The proposed IHPTDL-HAR technique aims at recogniz-ing the human actions in healthcare environment and helps the patients in mana-ging their healthcare service.In addition,the presented model makes use of Hierarchical Clustering(HC)-based outlier detection technique to remove the out-liers.IHPTDL-HAR technique incorporates DL-based Deep Belief Network(DBN)model to recognize the activities of users.Moreover,Harris Hawks Opti-mization(HHO)algorithm is used for hyperparameter tuning of DBN model.Finally,a comprehensive experimental analysis was conducted upon benchmark dataset and the results were examined under different aspects.The experimental results demonstrate that the proposed IHPTDL-HAR technique is a superior per-former compared to other recent techniques under different measures.

Faster Metallic Surface Defect Detection Using Deep Learning with Channel Shuffling

Deep learning has been constantly improving in recent years,and a significant number of researchers have devoted themselves to the research of defect detection algorithms.Detection and recognition of small and complex targets is still a problem that needs to be solved.The authors of this research would like to present an improved defect detection model for detecting small and complex defect targets in steel surfaces.During steel strip production,mechanical forces and environmental factors cause surface defects of the steel strip.Therefore,the detection of such defects is key to the production of high-quality products.Moreover,surface defects of the steel strip cause great economic losses to the high-tech industry.So far,few studies have explored methods of identifying the defects,and most of the currently available algorithms are not sufficiently effective.Therefore,this study presents an improved real-time metallic surface defect detection model based on You Only Look Once(YOLOv5)specially designed for small networks.For the smaller features of the target,the conventional part is replaced with a depthwise convolution and channel shuffle mechanism.Then assigning weights to Feature Pyramid Networks(FPN)output features and fusing them,increases feature propagation and the network’s characterization ability.The experimental results reveal that the improved proposed model outperforms other comparable models in terms of accuracy and detection time.The precision of the proposed model achieved by mAP@0.5 is 77.5%on the Northeastern University,Dataset(NEU-DET)and 70.18%on the GC10-DET datasets.

A novel extended Portuguese of Interactive and Multi-Criteria Decision Making and Archimedean Bonferroni mean operators based on prospect theory to select green supplier with complex q-rung orthopair fuzzy information

The word sustainable or green supply chain refers to the concept of incorporating sustainable environmental procedures into the traditional supply chain.Green supply chain management gives a chance to revise procedures,materials and operational ideas.Choosing the fuzziness of assessing data and the spiritual situations of experts in the decision-making procedure are two important issues.The main contribution of this analysis is to derive the theory of Archimedean Bonferroni mean operator for complex qrung orthopair fuzzy(CQROF)information,called the CQROF Archimedean Bonferroni mean and CQROF weighted Archimedean Bonferroni mean operators which are very valuable,dominant and classical type of aggregation operators used for examining the interrelationship among the finite number of attributes in modern data fusion theory.Inspirational and well-used properties of the initiated theories are also diagnosed with some special cases.Additionally,the theory of extended TODIM tool using the prospect theory based on CQROF information was discovered,which play an essential and critical role in the environment of fuzzy set theory.Finally,a real life problem by computing a green supply chain management based on the initiated CQROF operators was evaluated and fully illustrating the feasibility and efficiency of the diagnosed work with the help of a comparison between existing and prevailing theories.

具有适应性突变和惯性权重的粒子群优化(PSO)算法及其在动态系统参数估计中的应用(英文)

An important problem in engineering is the unknown parameters estimation in nonlinear systems.In this paper,a novel adaptive particle swarm optimization (APSO) method is proposed to solve this problem.This work considers two new aspects,namely an adaptive mutation mechanism and a dynamic inertia weight into the conventional particle swarm optimization (PSO) method.These mechanisms are employed to enhance global search ability and to increase accuracy.First,three well-known benchmark functions namely Griewank,Rosenbrock and Rastrigrin are utilized to test the ability of a search algorithm for identifying the global optimum.The performance of the proposed APSO is compared with advanced algorithms such as a nonlinearly decreasing weight PSO (NDWPSO) and a real-coded genetic algorithm (GA),in terms of parameter accuracy and convergence speed.It is confirmed that the proposed APSO is more successful than other aforementioned algorithms.Finally,the feasibility of this algorithm is demonstrated through estimating the parameters of two kinds of highly nonlinear systems as the case studies.

H_∞ Consensus Control of Discrete-Time Multi-Agent Systems Under Network Imperfections and External Disturbance

This paper presents a distributed control protocol for consensus control of multi-agent systems(MASs) under external disturbances and network imperfections, including communication delay and random packet dropout. To comply with the discrete nature of networked systems, in contrast to most of the existing work for MASs under network imperfections,the agents are modeled by discrete-time dynamics. The communication network is considered to be undirected, its delay is considered to be time-varying but bounded, and its packet dropout is modeled by a Bernoulli distributed white sequence.Sufficient conditions in terms of linear matrix inequalities(LMIs)for asymptotic mean-square consensus stability are derived under network imperfections without considering external disturbances.A desired disturbance attenuation level in the presence of both external disturbances and network imperfections is also provided.A simulation example is given to verify the effectiveness of the proposed approach in coping with network imperfection and disturbances.

Effect of heat input on microstructure and mechanical properties of butt-welded dissimilar magnesium alloys joint

The effects of heat input on the microstructures and mechanical properties of tungsten inert gas （TIG） butt-welded AZ31/MB3 dissimilar Mg alloys joint were investigated by microstructural observations, microhardness testing and tensile testing. The results reveal that with the increase of heat input, the width of welding seam increases obviously and the grains both in the fusion zone and the heat affected zone coarsen during TIG welding process. The tensile strength of butt-welded joint increases with the increase of heat input and the maximum joining strength of 242 MPa is obtained with wedding current of 90 A. However, lots of welding pores occur with the further increase of heat input, which results in the decrease of joining strength. It is experimentally demonstrated that robust joint can be obtained by TIG welding process.

Chaos detection and control in a typical power system

In this paper, a new chaotic system is introduced. The proposed system is a conventional power network that demonstrates a chaotic behavior under special operating conditions. Some features such as Lyapunov exponents and a strange attractor show the chaotic behavior of the system, which decreases the system performance. Two different controllers are proposed to control the chaotic system. The first one is a nonlinear conventional controller that is simple and easy to construct, but the second one is developed based on the finite time control theory and optimized for faster control. A MATLAB-based simulation verifies the results.

Parameter Identification Based on a Modified PSO Applied to Suspension System

This paper presents a novel modified particle swarm optimization algorithm (MPSO) for both offline and online parametric identification of dynamic models. The MPSO is applied for identifying a suspension system introduced by a quarter-car model. A novel mutation mechanism is employed in MPSO to enhance global search ability and increase convergence speed of basic PSO (BPSO) algorithm. MPSO optimization is used to find the optimum values of parameters by minimizing the sum of squares error. The performance of the MPSO is compared with other optimization methods including BPSO and Genetic Algorithm (GA) in offline parameter identification. The simulating results show that this algorithm not only has advantage of convergence property over BPSO and GA, but also can avoid the premature convergence problem effectively. The MPSO algorithm is also improved to detect and determine the variation of parameters. This novel algorithm is successfully applied for online parameter identification of suspension system.

Hybrid compensation method for traction power quality compensators in electrified railway power supply system

In order to improve the Power Quality(PQ)of traction power supply system and reduce the power rating and operation cost of compensator,a Static VAR Compensator(SVC)integrated Railway Power Conditioner(RPC)is presented in this paper.RPC is a widely used device in the AC electrified railway systems to enhance the PQ indices of the main network.The next generation of this equipment is Active Power Quality Compensator(APQC).The major concern of these compensators is their high kVA rating.In this paper,a hybrid technique is proposed to solve aforementioned problems.A combination of SVC as an auxiliary device is employed together with the main compensators,i.e.,RPC and APQC that leads on to the reduction of power rating of the main compensators.The use of proposed scheme will cause to reduce significantly the initial investment cost of compensation system.The main compensators are only utilized to balance active powers of two adjacent feeder sections and suppress harmonic currents.The SVCs are used to compensate reactive power and suppress the third and fifth harmonic currents.In this paper firstly,the PQ compensation procedure in AC electrified railway is analyzed step by step.Then,the control strategies for SVC and the main compensators are presented.Finally,a simulation is fulfilled using Matlab/Simulink software to verify the effectiveness and validity of the proposed scheme and compensation strategy and also demonstrate that this technique could compensate all PQ problems.

Wall Follower Autonomous Robot Development Applying Fuzzy Incremental Controller

This paper presents the design of an autonomous robot as a basic development of an intelligent wheeled mobile robot for air duct or corridor cleaning. The robot navigation is based on wall following algorithm. The robot is controlled using fuzzy incremental controller (FIC) and embedded in PIC18F4550 microcontroller. FIC guides the robot to move along a wall in a desired direction by maintaining a constant distance to the wall. Two ultrasonic sensors are installed in the left side of the robot to sense the wall distance. The signals from these sensors are fed to FIC that then used to determine the speed control of two DC motors. The robot movement is obtained through differentiating the speed of these two motors. The experimental results show that FIC is successfully controlling the robot to follow the wall as a guidance line and has good performance compare with PID controller.

Design of Fuzzy Controller for Robot Manipulators Using Bacterial Foraging Optimization Algorithm

Trial and error method can be used to find a suitable design of a fuzzy controller. However, there are many options including fuzzy rules, Membership Functions (MFs) and scaling factors to achieve a desired performance. An optimiza-tion algorithm facilitates this process and finds an optimal design to provide a desired performance. This paper presents a novel application of the Bacterial Foraging Optimization algorithm (BFO) to design a fuzzy controller for tracking control of a robot manipulator driven by permanent magnet DC motors. We use efficiently the BFO algorithm to form the rule base and MFs. The BFO algorithm is compared with a Particle Swarm Optimization algorithm (PSO). Performance of the controller in the joint space and in the Cartesian space is evaluated. Simulation results show superiority of the BFO algorithm to the PSO algorithm.

Machine Learning-Enabled Power Scheduling in IoT-Based Smart Cities

Recent advancements in hardware and communication technologies have enabled worldwide interconnection using the internet of things(IoT).The IoT is the backbone of smart city applications such as smart grids and green energy management.In smart cities,the IoT devices are used for linking power,price,energy,and demand information for smart homes and home energy management(HEM)in the smart grids.In complex smart gridconnected systems,power scheduling and secure dispatch of information are the main research challenge.These challenges can be resolved through various machine learning techniques and data analytics.In this paper,we have proposed a particle swarm optimization based machine learning algorithm known as a collaborative execute-before-after dependency-based requirement,for the smart grid.The proposed collaborative execute-before-after dependencybased requirement algorithm works in two phases,analysis and assessment of the requirements of end-users and power distribution companies.In the rst phases,a xed load is adjusted over a period of 24 h,and in the second phase,a randomly produced population load for 90 days is evaluated using particle swarm optimization.The simulation results demonstrate that the proposed algorithm performed better in terms of percentage cost reduction,peak to average ratio,and power variance mean ratio than particle swarm optimization and inclined block rate.

Ground demonstration system based on in-orbit assembly oriented manipulator flexible force control

To eliminate the load weight limit of carrier rockets and reduce the burden on support structures,in-orbit assembly is a key technology to make design of scattering a large diameter telescope into submirror modules,which requires smooth operation of assembly robots,and flexible force control technology is necessary. A ground demonstration system is presented for in-orbit assembly focusing on flexible force control. A six-dimensional force/torque sensor and its data acquisition system are used to compensate for gravity. For translation and rotation,an algorithm for flexible control is proposed. A ground transportation demonstration verifies accuracy and smoothness of flexible force control,and the transportation and assembly task is completed automatically. The proposed system is suitable for the development of in-orbit assembly robots.

Haptic display for virtual reality: progress and challenges

Immersion, interaction, and imagination are three features of virtual reality (VR). Existing VR systems possess fairly realistic visual and auditory feedbacks, and however, are poor with haptic feedback, by means of which human can perceive the physical world via abundant haptic properties. Haptic display is an interface aiming to enable bilateral signal communications between human and computer, and thus to greatly enhance the immersion and interaction of VR systems. This paper surveys the paradigm shift of haptic display occurred in the past 30 years, which is classified into three stages, including desktop haptics, surface haptics, and wearable haptics. The driving forces, key technologies and typical applications in each stage are critically reviewed. Toward the future high-fidelity VR interaction, research challenges are highlighted concerning handheld haptic device, multimodal haptic device, and high fidelity haptic rendering. In the end, the importance of understanding human haptic perception for designing effective haptic devices is addressed.

Wideband Cognitive Radio Networks Based Compressed Spectrum Sensing: A Survey

Spectrum sensing is a core function at cognitive radio systems to have spectrum awareness. This could be achieved by collecting samples from the frequency band under observation to make a conclusion whether the band is occupied, or it is a spectrum hole. The task of sensing is becoming more challenging especially at wideband spectrum scenario. The difficulty is due to conventional sampling rate theory which makes it infeasible to sample such very wide range of frequencies and the technical requirements are very costly. Recently, compressive sensing introduced itself as a pioneer solution that relaxed the wideband sampling rate requirements. It showed the ability to sample a signal below the Nyquist sampling rate and reconstructed it using very few measurements. In this paper, we discuss the approaches used for solving compressed spectrum sensing problem for wideband cognitive radio networks and how the problem is formulated and rendered to improve the detection performance.

Phase synchronization and synchronization frequency of two-coupled van der Pol oscillators with delayed coupling

In this paper, phase synchronization and the frequency of two synchronized van der Pol oscillators with delay coupling are studied. The dynamics of such a system are obtained using the describing function method, and the necessary conditions for phase synchronization are also achieved. Finding the vicinity of the synchronization frequency is the major advantage of the describing function method over other traditional methods. The equations obtained based on this method justify the phenomenon of the synchronization of coupled oscillators on a frequency either higher, between, or lower than the highest, in between, or lowest natural frequency of the aggregate oscillators. Several numerical examples simulate the different cases versus the various synchronization frequency delays.

Smart Anti-Pinch Window Simulation Using H_/H_(∞)Criterion and MOPSO

Automobile power windows are mechanisms that can be opened and shut with the press of a button.Although these windows can comfort the effort of occupancy to move the window,failure to recognize the person’s body part at the right time will result in damage and in some cases,loss of that part.An anti-pinch mechanism is an excellent choice to solve this problem,which detects the obstacle in the glass path immediately and moves it down.In this paper,an optimal solution H_/H_(∞)is presented for fault detection of the anti-pinch window system.The anti-pinch makes it possible to detect an obstacle and prevent damages through sampling parameters such as current consumption,the speed and the position of DC motors.In this research,a speed-based method is used to detect the obstacles.In order to secure the anti-pinch window,an optimal algorithm based on a fault detection observer is suggested.In the residual design,the proposed fault detection algorithm uses theDCmotor angular velocity rate.Robustness against disturbances and sensitivity to the faults are considered as an optimization problem based on Multi-Objective Particle Swarm Optimization algorithm.Finally,an optimal filter for solving the fault problem is designed using the H_/H_(∞)method.The results show that the simulated anti-pinch window is pretty sensitive to the fault,in the sense that it can detect the obstacle in 50 ms after the fault occurrence.

Design Methodology of a Micro-Scale 2-DOF Energy Harvesting Device for Low Frequency and Wide Bandwidth

A detailed design methodology of a micro-scale 2-DOF energy harvesting device that can harvest human motion energy of low frequency and wide bandwidth is developed. Based on the concept of the 2-DOF vibration absorber, device parameters are selected to harvest energy at low frequency of 1-10 Hz and wide bandwidth with ±20% of the mean frequency, which matches the human motion. The device dimensions are limited to 40 × 30 × 10 mm3 to fit with the human wrist size. Then, a finite element model is developed to investigate the system performance with the selected parameters. When subjected to harmonic excitation of 1 g, the proposed 2-DOF device is able to provide a power of at least 10 μW in between the two close resonant peaks of 4 Hz and 6 Hz, which is the target frequency range. The device shows very high power per square frequency compared with the reported harvesters.

Recurrent Autoencoder Ensembles for Brake Operating Unit Anomaly Detection on Metro Vehicles

The anomaly detection of the brake operating unit (BOU) in thebrake systems on metro vehicle is critical for the safety and reliability ofthe trains. On the other hand, current periodic inspection and maintenanceare unable to detect anomalies in an early stage. Also, building an accurateand stable system for detecting anomalies is extremely difficult. Therefore,we present an efficient model that use an ensemble of recurrent autoencodersto accurately detect the BOU abnormalities of metro trains. This is the firstproposal to employ an ensemble deep learning technique to detect BOUabnormalities in metro train braking systems. One of the anomalous caseson metro vehicles is the case when the air cylinder (AC) pressures are less thanthe brake cylinder (BC) pressures in certain parts where the brake pressuresincrease before coming to a halt. Hence, in this work, we first extract the dataof BC and AC pressures. Then, the extracted data of BC and AC pressuresare divided into multiple subsequences that are used as an input for bothbi-directional long short-term memory (biLSTM) and bi-directional gatedrecurrent unit (biGRU) autoencoders. The biLSTM and biGRU autoencodersare trained using training dataset that only contains normal subsequences. Fordetecting abnormalities from test dataset which consists of abnormal subsequences, the mean absolute errors (MAEs) between original subsequences andreconstructed subsequences from both biLSTM and biGRU autoencoders arecalculated. As an ensemble step, the total error is calculated by averaging twoMAEs from biLSTM and biGRU autoencoders. The subsequence with totalerror greater than a pre-defined threshold value is considered an abnormality.We carried out the experiments using the BOU dataset on metro vehiclesin South Korea. Experimental results demonstrate that the ensemble modelshows better performance than other autoencoder-based models, which showsthe effectiveness of our ensemble model for detecting BOU anomalies onmetro trains.

3D Instance Segmentation Using Deep Learning on RGB-D Indoor Data

3D object recognition is a challenging task for intelligent and robot systems in industrial and home indoor environments.It is critical for such systems to recognize and segment the 3D object instances that they encounter on a frequent basis.The computer vision,graphics,and machine learning fields have all given it a lot of attention.Traditionally,3D segmentation was done with hand-crafted features and designed approaches that didn’t achieve acceptable performance and couldn’t be generalized to large-scale data.Deep learning approaches have lately become the preferred method for 3D segmentation challenges by their great success in 2D computer vision.However,the task of instance segmentation is currently less explored.In this paper,we propose a novel approach for efficient 3D instance segmentation using red green blue and depth(RGB-D)data based on deep learning.The 2D region based convolutional neural networks(Mask R-CNN)deep learning model with point based rending module is adapted to integrate with depth information to recognize and segment 3D instances of objects.In order to generate 3D point cloud coordinates(x,y,z),segmented 2D pixels(u,v)of recognized object regions in the RGB image are merged into(u,v)points of the depth image.Moreover,we conducted an experiment and analysis to compare our proposed method from various points of view and distances.The experimentation shows the proposed 3D object recognition and instance segmentation are sufficiently beneficial to support object handling in robotic and intelligent systems.