
Intelligent Autonomous Systems 17
Description
Alles über E-Books | Antworten auf Fragen rund um E-Books, Kopierschutz und Dateiformate finden Sie in unserem Info- & Hilfebereich.
"IAS has been held every two years since 1986 providing venue for the latest accomplishments and innovations in advanced intelligent autonomous systems. New technologies and application domains continuously pose new challenges to be overcome in order to apply intelligent autonomous systems in a reliable and user-independent way in areas ranging from industrial applications to professional service and household domains. The present book contains the papers presented at the 17th International Conference on Intelligent Autonomous Systems (IAS-17), which was held from June 13-16, 2022, in Zagreb, Croatia. In our view, 62 papers, authored by 196 authors from 19 countries, are a testimony to the appeal of the conference considering travel restrictions imposed by the COVID-19 pandemic. Our special thanks go to the authors and the reviewers for their effort-the results of their joint work are visible in this book. We look forward to seeing you at IAS-18 in 2023 in Suwon, South Korea!"
More details
Other editions
Additional editions

Content
- Intro
- Program Committee
- Preface
- Contents
- Collaborative Robots
- Evaluation of Software Solutions for Risk Assessment Focusing on Human-Robot Collaboration
- 1 Introduction
- 2 Risk Assessment
- 3 Study Procedure
- 3.1 Selection of Software Solutions Included in the Study
- 3.2 Study Realization
- 3.3 Definition of Evaluation Criteria
- 3.4 Evaluation System and Weighting
- 4 Evaluation Results
- 4.1 Expert Requirements
- 4.2 Suitability for HRC
- 5 Discussion
- 6 Conclusion
- References
- Goal Inference via Corrective Path Demonstration for Human-Robot Collaboration
- 1 Introduction
- 2 Overview of the Proposed System
- 2.1 Assumed Environment
- 2.2 Flow of the Proposed System
- 3 Learning from Corrective Path Demonstration
- 3.1 Position Samples Generation
- 3.2 Learning Relations with Surrounding Objects Probabilistically
- 4 Goal Inference
- 4.1 Partial Probability Map Calculation
- 4.2 Integration and Goal Inference
- 5 Experiment
- 5.1 Experimental Setting
- 5.2 Result
- 6 Conclusion
- References
- Skeleton-Based Action and Gesture Recognition for Human-Robot Collaboration
- 1 Introduction
- 2 Related Works
- 3 Methods
- 3.1 Pose Estimation
- 3.2 3D Pose Estimation
- 3.3 Action and Gesture Recognition
- 3.4 Ensemble Averaging of the Classifiers Predictions
- 4 IAS-Lab Collaborative HAR Dataset
- 5 Experimental Results
- 5.1 Pre-training on the NTU RGB+D Dataset
- 5.2 Fine-Tuning on the IAS-Lab Collaborative HAR Dataset
- 5.3 Ensemble Results Using Body and Hands Models
- 6 Conclusions
- References
- Benchmarking Variable-Stiffness Grippers
- 1 Introduction
- 2 State of the Art
- 3 Test Description
- 3.1 Adaptation Test
- 3.2 Finger Force Test
- 3.3 Grasp Strength Test
- 3.4 Gripper Payload Test
- 4 Experiments
- 4.1 Variable-Stiffness Gripper (VSGrip)
- 4.2 Piston Gripper (PistGrip)
- 4.3 Parallel Gripper (ParGrip)
- 4.4 Stiffness States
- 5 Results and Discussions
- 5.1 Object Grasping Test
- 5.2 Finger Force Test
- 5.3 Grasp Strength Test
- 5.4 Gripper Payload Test
- 6 Conclusions
- References
- Evaluation of Safe Reinforcement Learning with CoMirror Algorithm in a Non-Markovian Reward Problem
- 1 Introduction
- 2 Related Work
- 3 Preliminaries and Problem Setting
- 4 CoMirror Descent Search
- 5 Experiments
- 5.1 Experiment with Non-Markovian Reward and Cost
- 5.2 Result
- 5.3 Discussion
- 6 Conclusion
- References
- Validation of Shared Intelligence Approach for Teleoperating Telepresence Robots Through Inaccurate Interfaces
- 1 Introduction
- 2 Shared Intelligence
- 3 Methods
- 3.1 Participants
- 3.2 User's Interfaces
- 3.3 Telepresence Platform
- 3.4 Experiment
- 4 Results
- 5 Conclusion
- References
- Uncertainty Estimation for Safe Human-Robot Collaboration Using Conservation Measures
- 1 Introduction
- 2 Uncertainty and Safety in HRC
- 3 Methodology
- 3.1 Problem Statement
- 3.2 Classification of Uncertainties
- 3.3 Uncertainty Quantification
- 3.4 Use Case: Human Robot Collaboration
- 4 Experiments
- 4.1 Datasets
- 4.2 Conservation Properties and Type B Uncertainties
- 4.3 Validation, Evaluation and Results
- 4.4 Results
- 4.5 Mapping on Safety Limit
- 4.6 Limitations
- 4.7 Discussion
- 5 Conclusion and Outlook
- References
- Post-facto Misrecognition Filter Based on Resumable Interruptions for Coping with Real World Uncertainty in the Development of Reactive Robotic Behaviors
- 1 Introduction
- 2 Past Work
- 2.1 Control Structures for Reactive Task Definition
- 2.2 Robust Recognition Using Neural Networks and Data Filtering Algorithms
- 2.3 Attaining Zero Resumption Overhead
- 3 Post-facto Misrecognition Filter Based on Resumable Interruptions
- 4 Experiments
- 4.1 Setup
- 4.2 Results
- 5 Conclusions
- References
- Gestural and Touchscreen Interaction for Human-Robot Collaboration: A Comparative Study
- 1 Introduction
- 2 Background
- 3 Design Principles
- 4 Implementation
- 4.1 Graphical User Interface
- 4.2 Input
- 4.3 Logic
- 5 Experiment
- 5.1 Experimental Setup
- 5.2 Description of the Experiment
- 6 Results
- 6.1 Assessment of SLOTH Performance
- 6.2 Analysis of Questionnaires
- 6.3 Time Analysis
- 7 Conclusions
- References
- Development and Evaluation of Fiber Reinforced Modular Soft Actuators and an Individualized Soft Rehabilitation Glove
- 1 Introduction
- 2 Methods
- 2.1 Soft Actuators
- 2.2 Dummy Finger and Hand
- 2.3 Rehabilitation Glove
- 2.4 Characterization of Soft Actuators with the Dummy Finger
- 2.5 Evaluation of Prototype Rehabilitation Glove
- 3 Results
- 3.1 Characterization of Actuators with the Dummy Fingers
- 3.2 Evaluation of Prototype Rehabilitation Glove
- 4 Discussion
- 5 Conclusion
- References
- Navigation
- Minimum Displacement Motion Planning for Movable Obstacles
- 1 Introduction
- 2 Minimum Displacement Motion Planning
- 3 Algorithm for Minimum Displacement Motion Planning
- 3.1 (a) Planning Phase
- 3.2 (b) Refinement Phase
- 3.3 Optimality
- 4 Examples and Applications
- 4.1 Theoretical Examples
- 4.2 Applications
- 5 Related Work
- 6 Conclusion
- References
- Mixed Use of Pontryagin's Principle and the Hamilton-Jacobi-Bellman Equation in Infinite- and Finite-Horizon Constrained Optimal Control
- 1 Introduction
- 2 Preliminaries
- 2.1 Motivations for Using DIDO
- 3 Problem Formulation
- 4 Methodology
- 4.1 Application to State- Constrained Problem
- 4.2 Weak Solutions to HJB
- 5 Solving the Original Problem
- 5.1 Convergence to Original Infinite-Horizon Problem
- 5.2 Approximation of Original Finite-Time Problem
- 6 Example
- 6.1 AUV
- 7 Conclusions and Future Work
- References
- Comparing SONN Types for Efficient Robot Motion Planning in the Configuration Space
- 1 Introduction
- 2 Self-Organizing Neural Networks
- 3 Methodology
- 3.1 Model Selection
- 3.2 Connection Reduction
- 4 Evaluation
- 4.1 Qualitative Analysis
- 4.2 Parameter Setting
- 4.3 Comparison of the Different Models
- 4.4 Path Analysis
- 5 Conclusion
- References
- Physics-Based Motion Planning of a Fruit Harvesting Manipulator for Pushing Obstacles in a Cluttered Environment
- 1 Introduction
- 2 Overview of the System
- 2.1 Work
- 2.2 Robot
- 2.3 System Flow
- 3 Physics-Based Motion Planning
- 3.1 Exploration in the Configuration Space
- 3.2 Physics Simulation
- 3.3 State Validity Check
- 3.4 Motion Selection and Reconstruction
- 4 Consideration of an Evaluation Index
- 5 Evaluation
- 5.1 Method
- 5.2 Conditions
- 5.3 Results
- 6 Conclusion
- References
- A Dynamics-Aware NMPC Method for Robot Navigation Among Moving Obstacles
- 1 Introduction
- 2 Problem Formulation
- 3 The Proposed NMPC Approach
- 4 Collision Avoidance
- 4.1 Preliminaries
- 4.2 Avoidable Collision States
- 4.3 Use of the ACS Condition in the NLP
- 5 Simulations
- 5.1 Static Environments
- 5.2 Dynamic Environments
- 6 Conclusion
- References
- Benchmark of Sampling-Based Optimizing Planners for Outdoor Robot Navigation
- 1 Introduction
- 2 Related Work
- 3 Problem Formulation
- 3.1 Evaluation Metrics
- 3.2 Evaluation Methodology
- 4 Experiments
- 4.1 Scenarios and Evaluation Setup
- 4.2 Interpreting Results Metric-Wise
- 4.3 Combining the Metrics for Absolute Ranking
- 5 Conclusion
- References
- 3D Traversability Analysis in Forest Environments Based on Mechanical Effort
- 1 Introduction
- 2 System Architecture
- 3 Experimental Evaluation
- 4 Results and Discussion
- 5 Conclusion
- References
- Two-step Planning of Dynamic UAV Trajectories Using Iterative -Spaces
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 -Space Definition
- 3.2 Iterative -Spaces
- 3.3 Application to UAV Trajectory Planning
- 4 Evaluation
- 5 Conclusion
- References
- End-to-End Path Estimation and Automatic Dataset Generation for Robot Navigation in Plant-Rich Environments
- 1 Introduction
- 2 Related Work
- 2.1 Traversable Region and Path Detection
- 2.2 Automatic Data Annotation
- 3 Overview of the Proposed Method
- 4 Automatic Data Generation by Trajectory Estimation
- 5 Path Estimation
- 6 Robot Control Based on the Estimated Path
- 7 Experiments
- 7.1 Training with the Automatically Generated Dataset
- 7.2 Navigation
- 8 Conclusions and Future Work
- References
- Localization on Indoor Topological Maps-SCAM: Scale-Compatible Adaptive Monte-Carlo Localization
- 1 Introduction
- 2 Background and Related Work
- 3 SCAM: Scale-Compatible Adaptive Monte-Carlo Localization
- 3.1 Prediction
- 3.2 Observation Update
- 4 Experimental Results and Discussion
- 4.1 Experimental Hardware and Set-up
- 4.2 Localization on Hand-Drawn Maps
- 4.3 Consistency Checks of Localization Results
- 4.4 Applicability of SCAM
- 4.5 Increased Initial Pose Error Tolerance
- 5 Conclusion
- References
- Perception
- Hyperspectral 3D Point Cloud Segmentation Using RandLA-Net
- 1 Introduction
- 2 Related Work
- 3 The Hyperspectral Semantic Street Scene Dataset
- 4 Semantic Classification with RandLA-Net
- 4.1 Pre-filtering
- 4.2 RandLA-Net Implementation and Parameterization
- 5 Evaluation
- 6 Conclusion and Outlook
- References
- 3D Semantic Scene Perception Using Distributed Smart Edge Sensors
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Smart Edge Sensor Hardware
- 3.2 Single-View Embedded Semantic Perception
- 3.3 Multi-modal Semantic Point Cloud Fusion
- 3.4 3D Semantic Mapping
- 3.5 3D Human Pose Estimation with Occlusion Feedback
- 4 Evaluation
- 4.1 Implementation Details
- 4.2 Quantitative Results
- 4.3 Qualitative Results
- 4.4 Run-time Analysis
- 5 Conclusion
- References
- Collision Warning by Rotating 2D LiDAR for Safe Crane Operation
- 1 Introduction
- 2 Sensor System Development and Preliminary Tests
- 3 Collision Warning
- 3.1 Approach
- 3.2 Determining the Condition on Scanning Angular Velocity scan
- 3.3 Determining Safety Margin rmargin
- 3.4 Determining Safety Margin for Collision of Moving Load and Static Obstacles
- 4 Experiments in Simulation
- 4.1 Crane and Sensor Simulator
- 4.2 Verifying Collision Warning in Simulation
- 5 Preliminary Real-World Experiments
- 5.1 Experimental Setup
- 5.2 Results
- 6 Conclusions and Future Work
- References
- Semantic Classification in Uncolored 3D Point Clouds Using Multiscale Features
- 1 Introduction
- 2 Random Forest Classification
- 2.1 Parameters in Our RF Implementation
- 2.2 Performance Evaluation Strategy
- 2.3 Handling of Highly Imbalanced Data
- 3 Tools and Datasets
- 4 Semantic Classification
- 4.1 Finding the Optimal Neighborhood Size
- 4.2 Definition of Multiscale Features
- 4.3 Related Work: Deep Learning for 3D Point Cloud Classification
- 5 Results and Comparison
- 5.1 Classification Results
- 5.2 Run Times
- 6 Conclusions and Future Work
- References
- On the Evaluation of RGB-D-Based Categorical Pose and Shape Estimation
- 1 Introduction
- 2 Related Work
- 3 Evaluation Protocol
- 3.1 Problem Definition
- 3.2 Metrics
- 3.3 Datasets
- 4 Experiments
- 5 Limitations
- 6 Conclusion and Outlook
- References
- On Hand-Eye Calibration via On-Manifold Gauss-Newton Optimization
- 1 Introduction
- 2 Mathematical Background
- 3 Proposed Method
- 4 Experimental Results
- 4.1 Simulation Results
- 4.2 Real World Data
- 5 Conclusion
- References
- YOLOPose: Transformer-Based Multi-object 6D Pose Estimation Using Keypoint Regression
- 1 Related Work
- 1.1 RGB Object Pose Estimation
- 1.2 Learned PnP
- 2 Method
- 2.1 Multi-object Keypoint Regression as Set Prediction
- 2.2 Keypoints Representation
- 2.3 RotEst
- 2.4 Loss Function
- 2.5 Model Architecture
- 3 Evaluation
- 3.1 Dataset
- 3.2 Metrics
- 3.3 Hyperparameters
- 3.4 Results
- 3.5 Inference Time Analysis
- 4 Ablation Study
- 4.1 Effectiveness of Keypoints Representations
- 4.2 Effectiveness of RotEst
- 5 Discussion & Conclusion
- References
- People Tracking in Panoramic Video for Guiding Robots
- 1 Introduction
- 2 Related Work
- 2.1 People Detection
- 2.2 People Tracking
- 3 Methodology
- 3.1 Baseline
- 3.2 The TILES Approach
- 3.3 The ROI Approach
- 3.4 People Tracking in Panoramic Videos
- 4 Experimental Evaluation
- 4.1 Experimental Setup
- 4.2 Evaluation Metrics
- 4.3 Results and Discussions
- 5 Conclusions and Future Work
- References
- Clustering-Based Refinement for 3D Human Body Parts Segmentation
- 1 Introduction
- 2 Related Works
- 2.1 Human Parsing
- 2.2 3D Human Parsing
- 2.3 Clustering Algorithms and Applications
- 3 Clustering-Based Algorithm for 3D Segmentation Refinement
- 3.1 Human Body Clusters Detection
- 3.2 Feature Extraction and Graph Construction
- 3.3 Segmentation Label Refinement
- 4 Experiments
- 4.1 Performance on Synthetic Data
- 4.2 Performance on Real Data
- 5 Conclusions
- References
- Autonomous Exploration for 3D Mapping Using a Mobile Manipulator Robot with an RGB-D Camera
- 1 Introduction
- 2 Related Works
- 3 Overview of the Proposed Exploration
- 4 Frontier Observation Map Generation
- 4.1 Frontier Detection
- 4.2 Frontier Observation Map Generation
- 5 Determination of Travel Destination and Arm Posture Control
- 5.1 Determination of Travel Destination
- 5.2 Moving to the Destination and Controlling the Arm Posture
- 6 Verifications
- 6.1 Verification Scenarios
- 6.2 Results and Discussion
- 7 Conclusion
- References
- Reaching Motion Planning with Vision-Based Deep Neural Networks for Dual Arm Robots
- 1 Introduction
- 2 Approach to Generate Target Object Images
- 3 Instance Segmentation Based on Mask R-CNN ch31Kaimingsps2017
- 4 Reaching Order Classifier Based on CNN
- 4.1 Architecture of CNN
- 4.2 Network Training
- 5 Reaching Motion Planning for Dual Arm Robot
- 5.1 Overview of Reaching Motion Control
- 5.2 Grasping Classifier Based on CNN
- 5.3 Reaching Motion Planner Based on CNN
- 6 Reaching Experiments for Multiple Objects
- 6.1 Settings
- 6.2 Results
- 7 Conclusions
- References
- Kinematic Calibration of a Collaborative Robot by a Marker Based Optical Measurement Procedure
- 1 Introduction
- 2 Hardware, Data Aquisition and Algorithms
- 2.1 Experimental Setting and Preparation
- 2.2 Camera Calibration
- 2.3 Referencing of the Coordinate System of the Robot
- 2.4 Measurement of Kinematics and Detection of Static Phases
- 3 Experiments
- 3.1 Determination of Joint Axes and DH-parameters
- 3.2 Moving the Robot Based on Optimized Modified DH-parameters
- 4 Results
- 5 Discussion
- 6 Conclusion and Future Work
- References
- Randomized Robotic Visual Quality Inspection with In-hand Camera
- 1 Introduction
- 2 Use-Case
- 3 Motion Planning
- 4 Experimental Evaluation
- 5 Discussion
- 6 Conclusion
- References
- Articulated Objects: From Detection to Manipulation-Survey
- 1 Introduction
- 2 Detection and Segmentation of Articulated Objects
- 3 Estimation of Joint Parameters and Configuration
- 4 Robot Manipulation with Articulated Objects
- 4.1 Manipulation for Estimation of Kinematic Parameters
- 4.2 Manipulation for Evaluation of Joint Parameters
- 5 Conclusion
- References
- Learning
- Learning Sequential Latent Variable Models from Multimodal Time Series Data
- 1 Introduction
- 2 Related Work
- 3 Methodology
- 3.1 Sequential Latent Variable Models
- 3.2 Multimodal Sequential Latent Variable Models
- 4 Experiments
- 4.1 Datasets
- 4.2 Network Architecture and Training
- 4.3 Image Prediction Experiments
- 4.4 Regression Experiments
- 5 Conclusions and Future Work
- References
- Pushing the Limits of Learning-Based Traversability Analysis for Autonomous Driving on CPU
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 Grid-Based Space Representation
- 3.2 Geometric-Based Feature Extraction
- 3.3 Appearance-Based Feature Extraction
- 3.4 Point Clouds Integration
- 3.5 Point Sorting into the Grid Cells
- 3.6 Outliers Filter
- 4 Experiments
- 4.1 SemanticKITTI Dataset
- 4.2 Traversability Ground Truth Extraction
- 4.3 Training and Parameters
- 4.4 Metrics
- 4.5 Comparison
- 4.6 Ablation Study
- 5 Conclusions
- References
- Sensor-Based Navigation Using Hierarchical Reinforcement Learning
- 1 Introduction
- 2 Related Work
- 3 Our Approach
- 3.1 Navigation Task
- 3.2 Preliminaries
- 3.3 Agent Structure
- 3.4 Reward Function
- 3.5 Reinforcement Learning
- 4 Experimental Evaluation
- 4.1 Setup
- 4.2 Top Layer Training
- 4.3 Comparative Evaluation
- 4.4 Transfer to Real-World
- 5 Conclusion
- References
- Traction Optimization for Robust Navigation in Unstructured Environments Using Deep Neural Networks on the Example of the Off-Road Truck Unimog
- 1 Introduction
- 2 Related Work
- 3 Off-Road Robot Unimog
- 4 Traction Opimization Concept Based on Surface Data
- 4.1 Off-Road Surface Data
- 4.2 Traction-Dependent Vehicle Configuration
- 5 Semantic Segmentation
- 5.1 Network Training
- 6 Experiments
- 6.1 Performance on Dataset
- 6.2 Performance on Vehicle Data
- 7 Conclusion
- References
- Flattening Clothes with a Single-Arm Robot Based on Reinforcement Learning
- 1 Introduction
- 2 Problem Description
- 2.1 State
- 2.2 Actions
- 2.3 Rewards
- 3 Methodology
- 3.1 Visual Perception
- 3.2 Neural Network Structure for DDPG
- 3.3 Simulation
- 4 Experiments
- 5 Discussion
- 6 Conclusions
- References
- Robot Simulations
- A Monte Carlo Framework for Incremental Improvement of Simulation Fidelity
- 1 Introduction
- 2 Related Work
- 3 Approach
- 3.1 Kernel Generation
- 3.2 Kernel Manager
- 4 Delivery Vehicle
- 5 Field Test
- 5.1 Design and Simulation
- 5.2 Description of the Field Trials
- 6 Results
- 6.1 Kernels Non-identity Transfer Functions
- 6.2 Kernels with Identity Transfer Functions
- 7 Conclusions
- References
- Real2Sim or Sim2Real: Robotics Visual Insertion Using Deep Reinforcement Learning and Real2Sim Policy Adaptation
- 1 Introduction
- 1.1 Minimum Infrastructure and Prerequisites
- 1.2 Real2Sim or Sim2Real
- 1.3 Contributions
- 2 Task Description
- 3 Related Works
- 3.1 Robotics Insertion
- 3.2 Reinforcement Learning and Sim2real Adaptation
- 4 Our Approach
- 4.1 Overview
- 4.2 Reinforcement Learning for Insertion Action Learning
- 4.3 Real2Sim and Sim2Real Domain Adaptation
- 4.4 Task Mixing, Randomization, and Vectorization
- 5 Experiment
- 5.1 Experiment Overview
- 5.2 Task Design
- 5.3 Convergence of Reinforcement Learning on Tasks
- 5.4 Mixing Learning
- 5.5 Test in Real Scenario and Real2Sim
- 5.6 Benchmark Comparison
- 5.7 Discussion and Future Work
- 6 Conclusion
- References
- Towards Synthetic Data: Dealing with the Texture-Bias in Sim2real Learning
- 1 Introduction
- 1.1 Related Work
- 1.2 Contributions
- 2 Method
- 2.1 Datasets
- 2.2 Semantic Segmentation Neural Network
- 3 Experimental Validation
- 3.1 Texture-Bias in Semantic Segmentation Model
- 3.2 Sim2real Approach for Leaf Segmentation in Robotic Harvesting
- 4 Conclusion
- References
- On Scene Engineering and Domain Randomization: Synthetic Data for Industrial Item Picking
- 1 Introduction
- 2 Related work
- 2.1 Cut and Paste
- 2.2 Domain Adaptation
- 2.3 Domain Randomization
- 3 Method
- 3.1 Data Generation
- 3.2 Training and Evaluation
- 4 Experiments
- 4.1 Base Scenes
- 4.2 Ablation Study: Scene Variations
- 5 Conclusions
- References
- Hybrid Intelligence for Visual Identity Design: A Case Study
- 1 Introduction
- 2 Related Work
- 3 Method
- 3.1 A Hand-Guided Letter "C"
- 3.2 An Artificially-Generated Letter "C"
- 3.3 A Brain Surgery Trajectory by RONNA, CRTA'S Flagship Robotic Innovation
- 4 Conclusion
- References
- Multi-robot Systems
- State-Aware Layered BTs-Behavior Tree Extensions for Post-Actions, Preferences and Local Priorities in Robotic Applications
- 1 Introduction
- 2 Past Work
- 2.1 Basic Syntax of Behavior Trees
- 2.2 Benefits of Behavior Trees Over Other Control Architectures
- 2.3 Behavior Tree Extensions for Robotics
- 2.4 Approaches for Handling Statelessness
- 3 State-Aware Layered Behavior Trees
- 3.1 Introducing State Operators
- 3.2 Introducing Priority Layers and Expanding the State Operator
- 4 Case Analysis
- 5 Conclusions
- References
- Multi-agent Coordination Based on POMDPs and Consensus for Active Perception
- 1 Introduction
- 2 Preliminaries
- 2.1 Partially Observable Markov Decision Processes
- 2.2 Consensus Protocol
- 3 Methodology
- 3.1 Measured Variables
- 3.2 Simulating Multiple Missions with Sensing Failures
- 4 Multi-agent Framework Based on the POMDP and Consensus
- 4.1 Single-agent POMDP Model
- 4.2 Novelty-Biased Consensus
- 4.3 Integration of Novelty-Biased Consensus and POMDPs
- 5 Results
- 5.1 Experimental Evaluation
- 6 Conclusion and Future Work
- References
- KI5GRob: Fusing Cloud Computing and AI for Scalable Robotic System in Production and Logistics
- 1 Introduction
- 2 State of the Art
- 2.1 Cloud Robotics
- 2.2 Cloud Based Motion Planning and Control
- 2.3 Robotic Manipulation and Transfer Learning
- 2.4 Sensor Data Processing
- 3 The KI5GRob Approach
- 3.1 Decoupled Robot Hardware
- 3.2 Microservice-Based Software Architecture
- 3.3 Software Deployment in Hybrid Environment
- 3.4 Motion Control as a Service
- 3.5 Robotic Manipulation
- 3.6 Multi-modal Perception
- 3.7 Transfer Learning
- 4 Conclusion
- References
- A Correlated Random Walk Model to Rapidly Approximate Hitting Time Distributions in Multi-robot Systems
- 1 Introduction
- 2 Problem Settings
- 3 Correlated Random Walk Computations
- 4 Verification by ARGoS
- 5 Discussion and Conclusions
- References
- Synthesis and Monitoring of Complex Tasks for Heterogeneous Robots in an Industry 4.0 Scenario
- 1 Introduction
- 2 Synthesis of State-Charts for Complex Tasks
- 2.1 State-Chart Generator Layer
- 2.2 State-chart and Observer Layer
- 2.3 Primitives Layer
- 3 Industry 4.0 Application
- 3.1 Observer
- 3.2 Primitives of the RB-Kairos 5
- 3.3 Primitives of the KUKA LBR Iiwa 14
- 3.4 Primitives ABB YuMi
- 4 Experimental Results
- 4.1 RB-Kairos
- 4.2 KUKA LBR Iiwa 14 and ABB Yumi for the Assembly Process
- 5 Conclusion
- References
- Automated Driving and Traffic
- Training Traffic Light Behavior with End-to-End Learning
- 1 Introduction
- 2 Related Work
- 2.1 Conditional Imitation Learning
- 2.2 Reinforcement Learning
- 3 Method
- 3.1 Network Architecture
- 3.2 PID Controller
- 4 Experimental Setup
- 4.1 Simulation Environment
- 4.2 Dataset
- 4.3 Evaluation
- 5 Results
- 5.1 Learning Curves
- 5.2 NoCrash results
- 6 Conclusion
- References
- On Demand Ride Sharing: Scheduling of an Autonoumous Bus Fleet for Last Mile Travel
- 1 Introduction
- 2 Related Work
- 3 Bus Fleet Scheduling
- 3.1 Representation of Street Network
- 3.2 Graph Generation
- 3.3 Optimization of Bus Trips Using Simulated Annealing
- 3.4 Construction and Solution of the ILP Problem
- 4 Experiments
- 5 Conclusion
- References
- Lane Change Classification with Neural Networks for Automated Conversion of Logical Scenarios
- 1 Introduction
- 2 Problem Description
- 3 Proposed Neural Network Models for Lane Change Classification
- 4 Training and Evaluation
- 4.1 Dataset Generation
- 5 Conclusion
- References
- Trajectory Analysis in a Lane-Based UAS Traffic Management System
- 1 Introduction and Background
- 1.1 Example ATOC Requirements by Utah Department of Transportation
- 2 Lane-Based Monitoring
- 2.1 NAB Modeling
- 2.2 NAB Analysis
- 3 Next Steps
- References
- Advanced and Intelligent Control Design for Underwater Robots
- Acoustical Underwater Localization of a Remotely Operated Vehicle in Mariculture
- 1 Introduction
- 2 Overview of Net Pen Inspection and Acoustical Underwater Localization Methods
- 3 Vehicles
- 3.1 Remotely Operated Underwater Vehicle (ROV)
- 3.2 Autonomous Surface Vehicle (ASV)
- 4 Underwater Acoustical Localization System
- 4.1 Acoustics Specifications
- 4.2 Mount Design
- 4.3 UWGPS-ROS2 Interface
- 5 Results
- 6 Conclusion
- References
- A Virtual Online Simulator Design for the Docking of Unmanned Underwater Vehicle
- 1 Introduction
- 2 Overall Framework of Virtual Online Simulator
- 3 Unit Design for Virtual Online Simulator
- 3.1 Virtual Docking Scene Design
- 3.2 Motion and Control Unit Design
- 3.3 Environmental Perception and Localization Unit Design
- 3.4 The Docking Navigation Unit Design
- 3.5 The Data Monitoring Unit Design
- 4 Application for the Underwater Tasks of UUV
- 4.1 The Motion Verification of UUV
- 4.2 The Underwater Docking of UUV
- 5 Conclusion
- References
- Simulation Environment for Underwater Vehicles Testing and Training in Unity3D
- 1 Introduction
- 2 Approach
- 3 Simulation Tests
- 4 Conclusions
- References
- Sliding Mode Control for Underwater Multi-DoF Hydraulic Manipulator
- 1 Introduction
- 2 Problem Formulation and Dynamic Models
- 3 Controller Design
- 4 Experiment Verification
- 4.1 Experimental Setup
- 4.2 Experimental Result
- 5 Conclusion
- References
- Path Following for Underwater Inspection Allowing Manoeuvring Constraints
- 1 Introduction
- 2 Modelling of the ROV
- 3 Path Planning
- 4 Path Following
- 4.1 Objective
- 4.2 Guidance Law
- 5 Control System
- 6 Simulation
- 6.1 Setup
- 6.2 Lawnmower Pattern
- 6.3 Inspection Pattern
- 7 Discussion
- 8 Conclusion
- References
- Autonomous Robots in Warehouse Automation
- Path Planning of Multiple Automatic Guided Vehicles with Tricycle Kinematics Considering Priorities and Occupancy Time Windows
- 1 Introduction
- 2 Path Planning of Multiple AGVs
- 2.1 Road Occupancy Check
- 2.2 Node Occupancy Check
- 2.3 Road and Node Occupancy Times for Tricycle Kinematics
- 2.4 Waiting Time Due to Occupancy
- 3 Experimental Set-up
- 3.1 Control of Multiple AGVs Along the Desired Paths
- 4 Results
- 4.1 Test Case 1: All the AGVs with the Same Speeds (Simulation)
- 4.2 Test Case 2: AGVs with Different Speeds (Simulation)
- 4.3 Test Case 3: AGVs with Different Speeds Considering Tricycle Kinematics (Simulation)
- 4.4 Test Case 4: AGVs with Different Speeds Considering Tricycle Kinematics (Real AGVs)
- 5 Conclusions
- References
- Cyber-Physical Platform with Miniature Robotic Vehicles for Research and Development of Autonomous Mobile Systems
- 1 Introduction
- 2 Small-Scale Physical Model of the Environment
- 2.1 Model of a Town
- 2.2 Model of an Industrial Hall
- 3 Miniature Robotic Vehicles
- 3.1 Miniature Wheeled Mobile Robot with Differential Drive
- 3.2 Miniature Automated Guided Vehicle
- 4 Global Vision-Based Object Tracking
- 5 Soft Sensors
- 6 Conclusions
- References
- Autonomous Hierarchy Creation for Path Planning of Mobile Robots in Large Environments
- 1 Introduction
- 2 Modeling of the Large Environment Using H-Graph
- 2.1 H-Graph Definition
- 2.2 Conditions for Optimality
- 3 Autonomous H-Graph Creation
- 3.1 Creating Occupancy Grid Map from the Floor Plan
- 3.2 Finding Narrow Passages, Bridge Nodes and Rooms
- 3.3 Creation of Higher Level Nodes
- 3.4 Finding Partial Paths
- 4 Simulation Results
- 4.1 H-Graph Analysis
- 4.2 H-Graph and Plain Graph Paths
- 4.3 Comparison of Partial Paths by E* and A*
- 5 Conclusion and Future Work
- References
- Improving the Flow in Multi-robot Logistic Systems Through Optimization of Layout Roadmaps
- 1 Introduction
- 2 Problem Formulation
- 3 Roadmap Generation Inspired by Ant Colony Optimization
- 4 Approach Validation
- 5 Discussion, Conclusions, and Future Work
- References
- Author Index
System requirements
File format: PDF
Copy protection: Watermark-DRM (Digital Rights Management)
System requirements:
- Computer (Windows; MacOS X; Linux): Use the free software Adobe Reader, Adobe Digital Editions, or any other PDF viewer of your choice (see eBook Help).
- Tablet/Smartphone (Android; iOS): Install the free app Adobe Digital Editions or another reading app for eBooks, e.g., PocketBook (see eBook Help).
- E-reader: Bookeen, Kobo, Pocketbook, Sony, Tolino and many more (only limited: Kindle).
The file format PDF always displays a book page identically on any hardware. This makes PDF suitable for complex layouts such as those used in textbooks and reference books (images, tables, columns, footnotes). Unfortunately, on the small screens of e-readers or smartphones, PDFs are rather annoying, requiring too much scrolling.
This eBook uses Watermark-DRM, a „soft” copy protection. This means that there are no technical restrictions to prevent illegal distribution. However, there is a personalised watermark embedded in the eBook that can be used to identify the purchaser of the eBook in the event of misuse and to provide evidence for legal purposes.
For more information, see our eBook Help page.