Artificial Intelligence (AI): Recent Trends and Applications

Cover
Half Title
Series Page
Title Page
Copyright Page
Table of Contents
Preface
Acknowledgements
Editor biographies
List of Contributors
Chapter 1 Advances in Large-Scale Systems Simulation Modelling Using Multi-Agent Architectures Optimized with Artificial Intelligence Techniques for Improved Concurrency-Supported Scheduling Mechanisms with Application to Wireless Systems Simulation
1.1 Literature Review
1.1.1 Simulation Methodologies Applied in Wireless Communication Systems (WCS)
1.1.1.1 Simulation of WCS
1.1.1.2 Discrete Event Simulation
1.1.1.3 Event Scheduling
1.1.2 Channel Assignment in WCS
1.1.3 Multi-Agent Systems in WCS
1.1.3.1 Agent and Multi-Agent Systems
1.1.3.2 Multi-Agent Systems in WCS
1.1.4 The Concept of Cellular Network
1.1.5 Simulation Languages (SLs)
1.2 The Proposed Simulation Model
1.2.1 Network Structure
1.2.1.1 Operational Parameters
1.2.2 Modelled Network Services and Channel Allocation
1.2.2.1 Network Services
1.2.2.2 Channel Allocation
1.2.2.3 Traffic Generation
1.2.3 The Multi-Agent/Multilayered Model
1.2.4 Theoretical Analysis of Agents Adapted to Modelled Network Services
1.2.4.1 Network Agent Definition
1.2.4.2 Architecture of the Intelligent Network Agents
1.2.4.3 Network Agent Interface
1.2.4.4 Network Agents Which Maintain State
1.2.4.5 Network Agent Utility Functions
1.2.4.6 Multi-Agent Encounters
1.2.5 Event Interleaving as Scheduling Technique Based on Real-Time Scheduling Theory
1.2.5.1 Real-Time Scheduling Algorithms for Implementing Synchronized Processes or Events
1.2.5.2 Process Life Span in a Real-Time Scheduling Set-Up
1.2.5.3 Scheduling Concurrent Events in WCS
1.2.5.4 Response Time Analysis
1.2.5.5 Pre-emptive Stationary Priority Scheduling (PSPS)
1.2.6 Supported DCA Variations
1.2.6.1 The Conventional Unbalanced Variation (Classical DCA)
1.2.6.2 The Conventional Balanced Variation (Min Cell Congestion)
1.2.6.3 The Conventional Best CNR Variation
1.2.6.4 The Conventional Round Blocking Variation
1.2.6.5 The Proposed Novel Artificial Intelligence Based Balanced and Best CNR DCA Variation for Concurrent Channel Assignment
1.2.7 Implementation Architectures
1.2.7.1 Conventional Model
1.2.7.2 Concurrent Models
1.3 Simulation Model Evaluation
1.3.1 Network Behaviour
1.3.2 Monte Carlo Simulation Method
1.3.3 Simulation Model Behaviour
1.3.4 Results Accuracy
1.3.5 Reference Analysis Model Employing One Cell Only
1.4 Experimental Results
1.4.1 Indicative Results Based on Five Days of Network Operation
1.4.2 Model Behaviour Based on Architectural Variations
1.4.3 Scheduling Mechanism Comparison
1.4.4 Response Time Analysis Results
1.5 Conclusions and Future Work
References
Chapter 2 Let’s Find Out: Why Do Users React Differently to Applications Infused with AI Algorithms?
2.1 Introduction
2.2 Related Work and Hypothesis Formulation
2.2.1 Excitement
2.2.2 Anger
2.2.3 Desire
2.2.4 Happiness
2.2.5 Relax
2.3 Methodology
2.3.1 Participants
2.3.2 Procedure
2.4 Findings
2.4.1 Descriptive Statistics and Hypothesis Testing Outcomes
2.4.2 Qualitative Feedback
2.5 Discussions
2.6 Limitations and Future Work
2.7 Conclusion
References
Chapter 3 AI vs. Machine Learning vs. Deep Learning
3.1 Introduction: Background and Driving Forces
3.2 Overview of Artificial Intelligence
3.3 Steps to Implement Artificial Intelligence Algorithms
3.4 When/Where/How/Why to Use Artificial Intelligence?
3.5 Examples for Artificial Intelligence Applications
3.6 Overview of Machine Learning
3.7 Steps to Implement Machine Learning Algorithms
3.8 When/Where/How/Why to Use Machine Learning?
3.9 Examples for Machine Learning Applications
3.10 Overview of Deep Learning
3.11 Steps to Implement Deep Learning Algorithms
3.12 When/Where/How/Why to Use Deep Learning?
3.13 Examples for Deep Learning Applications
3.14 Comparisons of Artificial Intelligence, Deep Learning, and Machine Learning
3.15 Summary
Chapter 4 AI and Big Data: Ethical Reasoning and Responsibility
4.1 Introduction
4.2 Ethics Reasoning in Artificial Intelligence
4.3 Ethical Responsibility in AI
References
Chapter 5 Online Liquid Level Estimation in Dynamic Environments Using Artificial Neural Network
5.1 Introduction
5.2 Liquid Level Measurement in Dynamic Environments
5.2.1 Influence of Temperature
5.2.2 Influence of Inclination
5.2.3 Influence of Sloshes
5.3 Sensor Design
5.3.1 Fibre Bragg Grating Sensor
5.3.2 Cantilever Beam
5.3.3 Float Sensor
5.3.4 System and Working Principle
5.4 Introducing Neural Networks for Accurate Level Prediction
5.4.1 Sampling of Sensor Output
5.4.2 Artificial Neural Networks
5.4.3 Activation Function
5.5 Wavelet Neural Network
5.5.1 Training of WNN
5.6 Results
5.7 Conclusion
References
Chapter 6 Computer Vision Concepts and Applications
6.1 Introduction
6.1.1 Evolution of Computer Vision
6.2 Feature Extraction
6.2.1 Types of Features
6.2.2 Feature Extraction Methods
6.2.2.1 I. Low-Level Features
6.2.2.2 Texture Estimator
6.2.2.3 Colour Histogram
6.2.2.4 Colour Descriptor
6.3 Object Detection
6.3.1 Image Classification
6.3.1.1 Classification and Localization
6.3.2 Image Segmentation
6.3.2.1 Semantic Segmentation
6.3.2.2 Demerits of Sliding Window
6.3.2.3 Instance Segmentation
6.3.3 Region-based Methods
6.3.3.1 Region Proposal
6.3.3.2 Region-based Convolutional Neural Network(R-CNN)
6.3.3.3 Fast Region-based Convolutional Neural Network
6.3.3.4 Faster Region-based Convolutional Neural Network
6.3.4 Alternative Methods
6.3.4.1 HOG Features
6.3.4.2 You Only Look Once (YOLO)
6.3.4.3 Demerits of YOLO
6.4 Computer Vision Hardware, Software, and Services
6.4.1 Computer Vision Hardware
6.4.2 Software Libraries and Tools
6.4.3 Computer Vision Services
6.5 Applications of Computer Vision
6.5.1 Healthcare
6.5.2 Augmented Reality
6.5.3 Vision-based Self-Driving Cars
6.5.4 Automatic Target Recognition and Detection
6.5.4.1 Case Study: Robotic Path Planning Using Visual Percepts
6.6 Conclusion and Future Directions
Bibliography
Chapter 7 Generative Adversarial Network: Concepts, Variants, and Applications
7.1 Introduction
7.2 Overview
7.2.1 Deep Learning
7.2.2 Deep Generative Models
7.2.3 Generative Adversarial Networks
7.3 GAN Architecture
7.3.1 General Structure
7.3.2 Adversarial Process
7.3.3 Background Mathematics
7.4 GAN Variations
7.4.1 Overview
7.4.2 Techniques
7.4.2.1 Architecture-based Variant Class
7.4.2.2 Formulation-based Variant Class
7.5 Applications
7.5.1 Image Generation and Prediction
7.5.2 Image Translation
7.5.3 Image Editing
7.5.4 3D Object Generation
7.5.5 Video Manipulation
7.5.6 Audio Generation and Translation
7.5.7 Medical Image Processing
7.6 Conclusion and Future Directions
Bibliography
Chapter 8 Detection and Classification of Power Quality Disturbances in Smart Grids Using Artificial Intelligence Methods
8.1 Introduction
8.1.1 Signal Processing (SP)-based PQD Detection Methods
8.1.2 Artificial Intelligent (AI) Methods for PQD Detection
8.2 Wavelet Transform (WT)-based PQD Detection Methods
8.2.1 Wavelet Transform (WT)
8.2.2 Proposed DWT-based PQD Detection Method
8.3 AI-based PQD Classification Methods
8.3.1 Deep Learning Structures
8.3.1.1 SAE-based Methods
8.3.1.2 DNN-based Methods
8.3.1.3 DBN (Deep belief network)-based Methods
8.3.1.4 CNN-based Methods
8.3.2 Proposed Deep Learning and WT-based Hybrid PQD Classification Method
8.4 Results
8.5 Conclusion
References
Chapter 9 Robust Design of Artificial Neural Network Methodology to Solve the Inverse Kinematics of a Manipulator of 6 DOF
9.1 Introduction
9.1.1 Kinematics of Robotic Manipulators
9.1.2 Artificial Neural Networks
9.1.3 Inverse Kinematics Solution with Artificial Neural Networks
9.1.4 Robust Design of Artificial Neural Networks
9.2 Robust Design of Artificial Neural Networks Methodology
9.3 Kinematics Analysis of Robotic Manipulator Called Ketzal
9.3.1 Data Set Description
9.3.2 Description of Reduction Data Filter Algorithm
9.3.3 Data Set Analysis of Training and Test
9.3.4 Planning and Experimentation Stage
9.3.5 Analysis and Confirmation Stage
9.4 Conclusions and Discussions
Future Scope
Acknowledgements
References
Chapter 10 Generative Adversarial Network and Its Applications
10.1 Introduction
10.2 Discriminative Learning vs. Generative Learning
10.3 Deep Generative Model
10.4 Variational Auto Encoders
10.5 Generative Adversarial Network
10.6 Architecture of Generative Adversarial Network
10.7 Variations of GAN Architectures
10.7.1 Fully Connected GAN (FCGAN)
10.7.2 Laplacian Pyramid of Adversarial Networks (LAPGAN)
10.7.3 Deep Convolutional GAN (DCGAN)
10.7.4 Conditional GAN
10.7.5 Least-Square GAN
10.7.6 Auxiliary Classifier GAN
10.7.7 InfoGAN
10.8 Applications of GAN
10.8.1 Image generation
10.8.2 Image Translation
10.8.3 Anomaly Detection
10.9 Conclusion
References
Chapter 11 Applications of Artificial Intelligence in Environmental Science
11.1 Introduction
11.1.1 Artificial Intelligence
11.1.2 Artificial Intelligence and Its Roles in the Environment
11.2 Technological Solutions
11.2.1 Autonomous and Connected Electric Vehicles
11.2.2 Conservation Biology
11.2.3 Next-Generation Weather and Climate Prediction
11.2.4 Smart Earth
11.3 AI in the Monitoring Environment
11.3.1 Monitoring Soil
11.3.2 Monitoring Water
11.3.3 Monitoring Air
11.4 Risks of Artificial Intelligence
11.4.1 Bias
11.4.2 Liability
11.4.3 ASI (Artificial Superintelligence)
11.5 Conclusion and Future
Acknowledgement
References
Chapter 12 A Genetic Algorithm-based Artificial Intelligence Solution for Optimizing E-Commerce Logistics Vehicle Routing
12.1 Introduction
12.2 Transport Costs and Goods Characteristics in Logistics
12.3 Logistics Challenges on Festive Days
12.4 AI in Logistics
12.5 Genetic Algorithms
12.6 Vehicle Routing Problem
12.7 Performance Measures
12.8 The Time-Dependent Vehicle Routing Problem with Time Windows
12.9 Application of Artificial Intelligence in Route Planning and Optimization
12.10 Development of GA for Time-Dependent Vehicle Routing Problem with Time Windows
12.10.1 The GA-based Crossover for the Time-Dependent VRPTW
12.10.2 Evaluation of Fitness Function
12.10.3 Mutation
12.11 Comparison of Algorithms in Terms of Vehicles Utilized
12.12 Conclusion
References
Chapter 13 Application of Machine Learning for Fault Detection and Energy Efficiency Improvement in HVAC Application
13.1 Introduction
13.2 Sustainable Living in Buildings
13.3 Energy Scenario in Buildings
13.4 Digitalization and Artificial Intelligence for Energy Efficiency in Buildings
13.4.1 Artificial Intelligence (AI)
13.4.2 Internet of Things (IoT)
13.4.3 Machine Learning (ML)
13.4.4 Influence of Digitalization on HVAC Systems
13.4.5 Energy Optimization and Scheduling
13.4.6 Predictive Maintenance and Fault Diagnosis
13.5 Conclusion
Conflict of Interest
References
Chapter 14 Smart City Using Artificial Intelligence Enabled by IoT
14.1 Introduction
14.2 Structure of a Smart City
14.3 Requirements to Build a Smart City
14.4 Augmented Reality and Virtual Reality in Building a Smart City
14.5 Major Components Required to Build a Smart City
14.6 Challenges in Building a Smart City
14.7 Technologies Involved in Building a Smart City
14.7.1 Integrating AI and IoT to Build a Smart City
14.7.2 Blockchain in Building a Smart City
14.7.3 Big Data in Building Smart City
14.7.4 Robotics in Building a Smart city
14.8 Components of a Smart City
14.8.1 Smart Energy
14.8.2 Smart Healthcare
14.8.3 Smart Traffic Management
14.8.4 Smart Parking
14.8.5 Smart Waste Management
14.8.6 Smart Lighting
14.8.7 Smart Governance
14.8.8 Smart Agriculture
14.9 Drawbacks in Implementing Smart Cities
14.10 Conclusion
References
Chapter 15 AI Emerging Communication and Computing
15.1 Introduction
15.2 Industrial Revolution 4.0
15.3 Stages of AI
15.4 Classification of AI
15.4.1 Artificial Narrow Intelligence
15.4.2 Artificial General Intelligence
15.4.3 Artificial Super Intelligence (ASI)
15.5 Machine Learning Algorithms
15.6 AI Emerging Communication
15.7 Conclusion
Bibliography
Index