Intelligent Autonomous Systems 15

Name: Intelligent Autonomous Systems 15 | Proceedings of the 15th International Conference IAS-15
Brand: Springer
Price: 213.99 EUR
Availability: OnlineOnly

Proceedings of the 15th International Conference IAS-15

Marcus Strand Rüdiger Dillmann Emanuele Menegatti Stefano Ghidoni(Editor)

Springer (Publisher)

1st Edition

Published on 31. December 2018

XVII, 985 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-01370-7 (ISBN)

€213.99incl. 7% vat

System requirements

for PDF with digital watermarking

E-Book Single Licence

Available for download

Description

More details

Other editions

Content

Intro
Preface
Organization
Program Committee
Contents
Path Planning, Localization and Navigation
A Viterbi-like Approach for Trajectory Planning with Different Maneuvers
Abstract
1 Introduction
2 Related Work
3 The Trajectory Planning
3.1 Basic Considerations
3.2 Primitive Trajectories and Maneuvers
3.3 Finding an Optimal Variation
3.4 Examples and Evaluation
4 Conclusions and Future Work
References
Improving Relaxation-Based Constrained Path Planning via Quadratic Programming
1 Introduction
2 Planning Algorithm
2.1 Differentiable Constraints
2.2 Local Motions Using Quadratic Programming
2.3 QPlan
3 Simulation Results
4 Conclusions
References
Robust Path Planning Against Pose Errors for Mobile Robots in Rough Terrain
1 Introduction
2 Approach
3 Error Estimation
3.1 Pose Error Estimation of Each Node
3.2 Collision Check Using Error Ellipsoid
4 Sampling Method of Velocity
5 Simulation
5.1 Simulation Setting
5.2 Simulation Result
6 Experiment
6.1 Experiment Setting
6.2 Experiment Result
7 Conclusion
References
Stereo Vision-Based Optimal Path Planning with Stochastic Maps for Mobile Robot Navigation
1 Introduction
2 Methodology
2.1 3D Point Cloud Model from Stereo Vision
2.2 Post Processing of 3D Point Cloud Model Phase
2.3 Building Navigation Map Phase
2.4 Scoring Of 2D Navigation Map Cells Phase
2.5 The Navigation Algorithm
3 Experimental Setup and Comparative Analysis of Test Results
4 Conclusion
References
Variance Based Trajectory Segmentation in Object-Centric Coordinates
1 Introduction
2 Trajectory's Variance of Human Demonstration
2.1 Preliminaries
2.2 Related Works
2.3 Paper Contribution
3 Task Segmentation and Motion Generation
3.1 Align Time Series Data
3.2 Task Segmentation
3.3 Motion Generation
4 Experiments with REAL ROBOT
5 Conclusion
References
MILP-Based Dual-Arm Motion Planning Considering Shared Transfer Path for Pick-Up and Place
Abstract
1 Introduction
2 MILP-Based Dual-Arm Motion Planning Submission
2.1 Virtual Objects and Goals
2.2 Formulation of Pick-Up and Place Work
2.3 Path Planning with RRT
3 Result Simulation
4 Experiment
4.1 Experiment 1
4.2 Collision Avoidance Using Each Arm
5 Conclusion
Acknowledgment
References
Multiple Path Planner Integration for Obstacle Avoidance: MoveIt! and Potential Field Planner Synergy
1 Introduction
2 MoveIt! and OMPL
3 Potential Field Planner
4 Synergy Protocol
5 Experiments
6 Results and Conclusions
7 Code Repository
References
An Integrated Planning of Exploration, Coverage, and Object Localization for an Efficient Indoor Semantic Mapping
1 Introduction
2 Related Work
2.1 Exploration
2.2 Coverage
2.3 Object Search
2.4 Integrated Viewpoint Planning
3 Target Task and Overview of Integrated Viewpoint Planning
4 Integration of Exploration and Mapping Viewpoint Planning
4.1 Viewpoint Generation for Exploration
4.2 Viewpoint Generation for Coverage
4.3 Integrated Planning
5 Object Verification and Viewpoint Generation
5.1 Object Candidate Detection and Verification
5.2 Verification Viewpoint Generation
5.3 Integrated Planning
6 Experimental Results
6.1 Determining the Threshold for the Increment of Free Space Area
6.2 Mapping Results and Comparison Among Strategies
7 Conclusions and Future Work
References
Unsupervised Hump Detection for Mobile Robots Based On Kinematic Measurements and Deep-Learning Based Autoencoder
1 Introduction
2 Concept
3 Anomaly Detection and Deep Learning
4 Data Acquisition
4.1 Experimental Measurement Settings and Preparation
4.2 Measurement of Kinematics
4.3 Preprocessing and Visualization of Kinematic Data
5 Experiments
5.1 RNN Architecture
5.2 Training and Anomaly Detection
6 Results
6.1 Kinematic Measurements
6.2 Anomaly Detection
7 Discussion
8 Conclusion and Future Work
References
Efficient Coverage of Unstructured Environments
1 Introduction
2 Related Work
3 Problem Formulation
4 Our Approach to Coverage Planning
4.1 Decomposition of the Environment
4.2 Possible Boustrophedon Paths
4.3 Formulation as Traveling Salesman Problem
4.4 Solving the Traveling Salesman Problem
5 Considering Capacity Constraints
6 Path Execution and Replanning
7 Experimental Results
7.1 Comparative Experiments
7.2 Performance in Dynamic Scenarios
8 Conclusions
References
Intelligent Systems
Probabilistic Logic for Intelligent Systems
1 Introduction
2 Related Work
3 Method
3.1 When Variables Are Not Independent
4 Experiments and Results
4.1 Independent Variables
4.2 Non-independent Variables
4.3 Trajectory Visualization and Finding Good Initial Guesses
5 Conclusions and Future Work
5.1 Numerical Solutions
References
Aggregating Models for Anomaly Detection in Space Systems: Results from the FCTMAS Study
1 Introduction
2 The Application Context
3 Related Work
4 Problem Formulation
5 Problem Analysis and Solution
5.1 Analysis of the State Space
5.2 Solving Algorithms
6 Experimental Evaluation
7 Conclusion
References
Detection of Motion Patterns and Transition Conditions for Automatic Flow Diagram Generation of Robotic Tasks
1 Introduction
2 Visualization of Source Code
2.1 Schematic Representations
2.2 Our Method
3 Detecting Motion Patterns in Robotic Tasks
3.1 Code Segmentation
3.2 Pattern Detection and Regrouping
4 Detecting Transition Conditions
4.1 Identifying Triggering Clauses
4.2 Determining Relations Between Motion Patterns
5 Experiment on Flow Diagram Generation
6 Conclusion
References
Least Action Sequence Determination in the Planning of Non-prehensile Manipulation with Multiple Mobile Robots
1 Introduction
2 Methodology
2.1 Generation of Manipulation States
2.2 Transition Between Manipulation States
3 Implementation
4 Conclusion
References
Convolutional Channel Features-Based Person Identification for Person Following Robots
1 Introduction
2 Related Work
3 Person Tracking
3.1 LRF-Based Person Tracking
3.2 Visual Person Region Detection
4 Person Identification
4.1 Convolutional Channel Features
4.2 Online Boosting-Based Person Classifier
4.3 Evaluation
5 Person Following Experiment
6 Conclusion
References
Service Robot Using Estimation of Body Direction Based on Gait for Human-Robot Interaction
Abstract
1 Introduction
2 System Overview
3 Gait Measurement Algorithm for Moving Robot
3.1 Human Walking Model
3.2 State Equation of Each Leg
3.3 Gait Phase Judgement
3.4 Leg Tracking Considering Gait Phase
4 Estimation of Body Direction Based on Gait
4.1 Observation of Body Pose Based on Gait
4.2 State Equation of Body
5 Experiments
6 Conclusion
Acknowledgement
References
Accurate Pouring with an Autonomous Robot Using an RGB-D Camera
1 Introduction
2 Related Work
3 The Autonomous Pouring System
3.1 Liquid Height Detection
3.2 Kalman Filter
3.3 Pouring Control
4 Experiments
4.1 Influence of Different Liquids
4.2 Influence of Varying Initial Volume
4.3 Influence of Varying Target Heights
4.4 Influence of Bottle Opening
4.5 Influence of Different Cups
4.6 Influence of Initial Liquid in Cup
4.7 Comparison of Pouring Accuracy
5 Conclusion
References
MS3D: Mean-Shift Object Tracking Boosted by Joint Back Projection of Color and Depth
1 Introduction
2 Related Work
3 Proposed Algorithm
3.1 Discriminative Color Model
3.2 Back Projection of Color and Depth PDF
3.3 Occlusion Aware Mean-Shift Tracking
4 Experiment Results
5 Conclusion
References
Heterogeneous Multi-agent Routing Strategy for Robot-and-Picker-to-Good Order Fulfillment System
1 Introduction
2 Routing Problems Formulation
2.1 Robot Routing Problem Formulation
2.2 Worker Routing Problem Formulation
3 Genetic Algorithm Based Optimization
4 Simulations and Results
4.1 Description of Warehouse, Order Batches and Agents
4.2 Simulation Settings and Algorithmic Parameters
4.3 Results, Analysis and Solution Comparison
5 Conclusions and Future Work
References
Multiple Mobile Robot Management System for Transportation Tasks in Automated Laboratories Environment
Abstract
1 Introduction
2 System Structure
3 Communication
4 Robot Selection Strategies
4.1 The Highest Charging Value
4.2 Nearest Robot Working Strategy
5 Robot-Robot Interaction
6 Experiment
7 Conclusion
Acknowledgment
References
Robot Design
A Rolling Contact Joint Lower Extremity Exoskeleton Knee
1 Introduction
2 Requirements
3 Design
3.1 Modelling of the Rolling Contact Joint
3.2 Mechanical Design
3.3 Actuation
4 Evaluation and Results
4.1 Prototype
4.2 Experiments
5 Conclusion and Future Work
References
Towards a Stair Climbing Robot System Based on a Re-configurable Linkage Mechanism
Abstract
1 Introduction
2 Related Work
2.1 Stair Climbing Robots
2.2 Re-configurable Leg Mechanisms
3 Simulated Model
4 Mechanical Design and Fabrication
4.1 Full Robot Design Concept
4.2 CAD Model
4.3 Prototype Fabrication
5 Experimental Results
6 Conclusions and Future Work
References
Can Walking Be Modeled in a Pure Mechanical Fashion
1 Introduction
1.1 Legged Locomotion Survey
2 Problem Position
2.1 Spring-Loaded Inverted Pendulum Motion
2.2 Energy-Based Motion
2.3 Balanced Mode
3 Gait Pattern
3.1 Motivations for an Elliptic Model
3.2 Maintaining the Gait Cycle
4 Conclusions
References
Modular Robot that Modeled Cell Membrane Dynamics of a Cellular Slime Mold
1 Introduction
2 Self-reconfigurable Cellular Slime Mold
2.1 Cellular Slime Mold Deformation by Cell Membrane Dynamics
2.2 Cell Membrane Deformation by Pseudopod Elongation
3 Chemical Reaction Model for Deformation
3.1 The Phosphatidylinositol Lipids Signaling System
3.2 Cell Cytoskeleton Polymerization Model
4 Development of Real Physical Modular Robots
4.1 Overview of the Real Physical Modular Robots
4.2 Actuation of Connected Modular Robots
4.3 Control System
5 Verification of Self-reconfiguration
5.1 Experimental Setting
5.2 Experimental Results
6 Conclusion and Future Work
References
Locomotion of Hydraulic Amoeba-Like Robot Utilizing Transition of Mass Distribution
1 Introduction
2 Configuration of the Robot
2.1 Configuration of the Robot
2.2 Pipe and Pump Arrangement
2.3 Slippery Nylon Net for Effective Inflation of Fore-Chamber
3 Realization of Locomotion
3.1 Procedure for Locomotion
3.2 Realization of the Locomotion
4 Experiment
5 Conclusion
References
Common Dimensional Autoencoder for Identifying Agonist-Antagonist Muscle Pairs in Musculoskeletal Robots
1 Introduction
2 Proposed Method
2.1 Linear Model Extraction
2.2 Common Dimensional Autoencoder
3 Simulation
3.1 Setup
3.2 Results
4 Discussion and Conclusion
References
Analysis of Variable-Stiffness Soft Finger Joints
1 Introduction
2 Fingers
2.1 Finger Design
2.2 Prototype Manufacturing
3 Testing Bench
3.1 Making the Structure
3.2 Sensors
4 Experimental Testing of the Fingers
4.1 Test Description
4.2 Test Implementation
5 Results
5.1 Analogical Reading
5.2 Calculation of the Real Values
5.3 Discussion
6 Conclusions and Future Work
References
Method for Robot to Create New Function by Uniting with Surrounding Objects
1 Introduction
2 Related Works
3 Robot that Creates New Functions by Uniting with Surrounding Objects
3.1 Application
3.2 Necessary Factor of the Robot that Creates Function by Uniting with Surrounding Objects
4 First Prototype and Demonstration
4.1 First Prototype
4.2 Demonstration
5 Second Prototype and Demonstrations
5.1 The Second Prototype
5.2 Demonstration of Creating a Function to Handle Object on a High Place
5.3 Demonstration of Creating Automatic Open and Close Function for Door
6 Conclusion and Future Works
References
Sensing and Actuation
Configuration Depending Crosstalk Torque Calibration for Robotic Manipulators with Deep Neural Regression Models
1 Introduction
2 Related Work
3 Problem Description
4 Methodology
4.1 Random Forest Regression
4.2 Neural Network Regression
4.3 Lookup Table
5 Experiments
5.1 Data Acquisition
5.2 Evaluation of Lookup Tables
5.3 Evaluation of NN and RF Regression
6 Result/Discussion
7 Conclusion
References
Simulation of the SynTouch BioTac Sensor
1 Introduction
2 Related Work
3 The BioTac Sensor
3.1 Raw Sensor Data
3.2 Experimental Setup
3.3 Captured Dataset and Postprocessing
4 Simulation Model
4.1 Neural Network Architecture
4.2 Network Inputs and Outputs
4.3 Implementation
5 Evaluation
5.1 Performance
6 Conclusions
References
Force Sensing for Multi-point Contact Using a Constrained, Passive Joint Based on the Moment-Equivalent Point
1 Introduction
2 Moment-Equivalent Point
3 Force Sensing Based on MEP
4 Experiment
4.1 Experimental Setup
4.2 Experimental Procedure
4.3 Results
5 Conclusion
References
Simulation and Transfer of Reinforcement Learning Algorithms for Autonomous Obstacle Avoidance
1 Introduction
2 Methods
2.1 Simulation
2.2 Transfer
3 Results
3.1 Simulation
3.2 Transfer
4 Discussion
References
HI-VAL: Iterative Learning of Hierarchical Value Functions for Policy Generation
1 Introduction
2 Related Work
3 HI-VAL
3.1 Exploration and Sample Collection
3.2 Hierarchical Action Selection
3.3 HI-VAL Algorithm
4 Experimental Evaluation
4.1 Fetching Task
4.2 Pick and Delivery Task
5 Conclusion
References
Learning-Based Task Failure Prediction for Selective Dual-Arm Manipulation in Warehouse Stowing
1 Introduction
2 Related Work
3 Overview
3.1 Stowing Task and Task Failures
3.2 Selective Dual-Arm Stowing System
4 Selective Dual-ARM Stowing with Failure Prediction
4.1 Dual-Arm Stowing Motion
4.2 Task Failure Prediction
4.3 Motion Select Algorithm
4.4 Motion Select System
5 Experiments
5.1 Stowing Task Dataset
5.2 Task Failure Prediction
5.3 Selective Execution of Dual-Arm Support Motion
6 Conclusion
References
Learning of Motion Primitives Using Reference-Point-Dependent GP-HSMM for Domestic Service Robots
1 Introduction
2 Discovery of Motion Primitives
2.1 Segmentation of Time-Series Data
2.2 Reference-Point-Dependent Hidden Semi-Markov Model
3 Design of Reference Point
3.1 Daily Action in the Home
3.2 Reference Points for Expressing Trajectory
4 Experiment
4.1 Discovered Motion Primitives
5 Conclusion
References
Nonlinear Model Predictive Control for Two-Wheeled Service Robots
Abstract
1 Introduction
2 Two-Wheeled Service Robots
3 Control System Design for Two-Wheeled Service Robots
3.1 Model
3.2 Setting up the Goal
3.3 Nonlinear Model Predictive Control Problem
3.4 Setting the Evaluation Function Weight
3.5 Stability of Proposed System
4 Simulation
4.1 Simulation Conditions
4.2 Simulation Results
5 Experiments
6 Conclusion
References
A Generalised Method for Adaptive Longitudinal Control Using Reinforcement Learning
1 Introduction
2 Problem Formulation
3 Approach
3.1 Generalised Approach to Longitudinal Control
3.2 Least Square Policy Iteration
3.3 Learning the ACC Through LSPI
3.4 Cruise Control Mode
3.5 ACC and Cruise Control Modes Combined
4 Implementation
4.1 Experimental Setup
5 Experiments
5.1 Single-Car Scenario: single-car
5.2 Multi-cars Scenario: multi-car
6 Conclusion
References
Reconstructing State-Space from Movie Using Convolutional Autoencoder for Robot Control
1 Introduction
2 Proposed Method
2.1 State-Space Reconstruction
2.2 Linear Model Extraction and Control
3 Simulation
4 Discussion and Conclusion
References
BSplines Properties with Interval Analysis for Constraint Satisfaction Problem: Application in Robotics
1 Introduction
2 Problem Statement
3 Interval Analysis
3.1 Presentation
3.2 Solving CSP Using Interval Analysis
3.3 The Pessimism
4 BSplines Identification
4.1 Definition and Convex Hull Property
4.2 Multi-dimension BSplines
4.3 Control Point Identification
4.4 Kronecker Product
4.5 Properties of the Kronecker Product
4.6 Recursive Inverse Kronecker Product
4.7 Example
5 BSplines and Kronecker Product in the Contraction
5.1 One Constraint Contraction
5.2 Monomial Contraction
5.3 Implementation Trick
5.4 How to Deal with Non Linear Functions?
6 Simulations and Results
7 Conclusion and Perspectives
References
Human Detection and Interaction
Robot Vision System for Real-Time Human Detection and Action Recognition
1 Introduction
2 Input Image Based on Optical Flow
3 Convolutional Neural Network
4 Local Search Window
5 Experiment
5.1 Settings
5.2 Classifier for Humans
5.3 Classifier for Actions
5.4 Effectiveness of Local Search Window
5.5 Real-Time Human Detection and Action Recognition
6 Conclusions and Future Works
References
Movement Based Classification of People with Stroke Through Automated Analysis of Three-Dimensional Motion Data
Abstract
1 Introduction
2 Methods
2.1 Subjects
2.2 Tasks
2.3 Data Collection
2.4 Data Reduction
2.5 Feature Extraction
2.6 Data Classification
3 Results
4 Discussion
5 Conclusion
Acknowledgements and Funding
References
Real-Time Marker-Less Multi-person 3D Pose Estimation in RGB-Depth Camera Networks
1 Introduction
2 Related Work
2.1 Single-View Body Pose Estimation
2.2 Multi-view Body Pose Estimation
3 System Design
3.1 Camera Network Setup
3.2 Single-View Estimation of 3D Poses
3.3 Multi-view Fusion of 3D Poses
4 Experiments
4.1 Implementation Details
5 Conclusions and Future Works
References
People Finding Under Visibility Constraints Using Graph-Based Motion Prediction
1 Introduction
2 Related Work
3 Problem Formulation
4 Graph-Based People Tracking
5 Selecting Search Locations via Monte Carlo Simulations
6 Experimental Results
6.1 Experimental Setup
6.2 Evaluation and Results
7 Conclusion
References
Proposal and Validation of an Index for the Operator's Haptic Sensitivity in a Master-Slave System
Abstract
1 Introduction
2 Proposal of the Haptic Sensitivity Index
2.1 Dynamic Factors in Perception
2.2 Definition of the Index
3 Properties of the Haptic Sensitivity Index
3.1 Experimental Apparatus
3.2 Subject and Motion
3.3 Experiment Introduction
3.3.1 Objective
3.3.2 Experimental Procedure
3.3.3 Experiment Results
4 Validity of the Haptic Sensitivity Index
4.1 Relationship Between Parameters and C Value
4.2 Experimental Procedure
4.3 Experimental Results
4.3.1 Pairs of Parameter Sets with the Same Index Value
4.3.2 Pairs of Parameter Sets with just Noticeably Different Index Values
5 Discussion and Future Works
5.1 Relationship Between Haptic Sensitivity and Contact Velocity
5.2 Application of the Index
5.3 Future Works
References
Operating a Robot by Nonverbal Voice Expressed with Acoustic Features
1 Introduction
2 Nonverbal Voice for Operating Robots
2.1 Related Work
2.2 Using Characteristic of Frequency Domain for Operating a Robot
3 Multiplexing Operation by Nonverbal Voice
3.1 Three Kinds of Voice for the Operation
3.2 Possibilities of the Utterance
3.3 Grasping Experiments
4 Operation with Tongue Positions Distinguished by Nonverbal Voice
4.1 Distinguishing Tongue Positions by Nonverbal Voice
4.2 Grasping Experiments
5 Conclusion
References
Reduced Feature Set for Emotion Recognition Based on Angle and Size Information
1 Introduction
2 Problem Statement
3 Feature Creation
3.1 Generalized Procrustes Analysis
3.2 Angle and Size Features
4 Experiments
5 Conclusion
References
Multimodal Path Planning Using Potential Field for Human-Robot Interaction
Abstract
1 Introduction
2 Overview of Robot System
3 Recognition Reliability
3.1 Human Recognition Method
3.2 Recognition Reliability r_{All}^{I}
3.3 Experiment for Evaluating Recognition Reliability Characteristics
3.4 Formulation of Recognition Reliability Characteristics r_{{}}^{I}
4 Multimodal Path Planning
4.1 Design of Each Potential
4.2 Weighting Each Potential
4.3 Path Planning Considering the Potential Field
5 Experiment
6 Conclusion
Acknowledgment
References
2D and 3D Computer Vision for Robotics
Deep Learning Waterline Detection for Low-Cost Autonomous Boats
1 Introduction
2 Related Work
3 Methods
3.1 Network Architecture
3.2 Data Set
3.3 Training
3.4 Waterline Detection Algorithm
4 Experimental Results
5 Conclusions
References
Context-Aware Recognition of Drivable Terrain with Automated Parameters Estimation
1 Introduction
2 Related Work
3 Terrain Classification System
3.1 System Structure
3.2 Local Terrain Representation
3.3 Extraction of Features
3.4 Classification
4 Probabilistic Inference
5 Estimation of Parameters
6 Experiments and Results
7 Conclusions
References
Crop Edge Detection Based on Stereo Vision
1 Introduction
2 Stereo Vision System
3 Height Segmentation
3.1 Motivation
4 Elevation Grid Map
4.1 Expectation Maximization with Gaussian Mixture Model
4.2 Bayesian Filtering
5 Crop Edge Detection
6 Linear Crop Edge Model
6.1 Motivation
6.2 Model Fitting
6.3 Model Crop Edge Filtering
6.4 Model Confidence
7 Experiments and Results
8 Conclusions
References
Extracting Structure of Buildings Using Layout Reconstruction
1 Introduction
2 Related Work
3 Our Method
4 Experimental Results
4.1 Results with Grid Maps from Stage
4.2 Results with Publicly Available Datasets
4.3 Results with Partial Grid Maps
4.4 Results with Blueprints and Evacuation Maps
4.5 Results with a Grid Map Acquired by a Real Robot
5 Conclusions
References
Unknown Object Detection by Punching: An Impacting-Based Approach to Picking Novel Objects
1 Introduction
2 Overview of Impacting-Based Object Detection
3 Details of Object Detection
3.1 Keypoint Detection
3.2 Keypoint Matching
3.3 Keypoint Pair Filtering
3.4 Object Tracking
3.5 Object Segmentation
3.6 Picking
4 Experiments
4.1 Experimental Setup
4.2 Assumptions
4.3 Experiments of Object Detection
4.4 Picking Experiments
5 Conclusion
References
Efficient Semantic Segmentation for Visual Bird's-Eye View Interpretation
1 Introduction
2 Related Work
3 Methods
3.1 ArgMax Implementation
3.2 ENet's Channel Pruning
4 Channel Factor Selection
5 Results
6 Conclusion
References
Concept Study for Vehicle Self-Localization Using Neural Networks for Detection of Pole-Like Landmarks
1 Introduction
2 Related Work
3 Overview of the Approach
4 Implementation
4.1 Perception via LiDAR
4.2 Perception via Stereo Camera
4.3 Perception via Mono Camera
5 Results
5.1 Old Approach
5.2 New Approach
6 Conclusion
References
Cross Domain Image Transformation Using Effective Latent Space Association
1 Introduction
2 Related Work
3 Model
4 Training Details
5 Experiments
5.1 Results
6 Conclusion
References
Global Registration of Point Clouds for Mapping
1 Introduction
1.1 Related Work
1.2 Overview
2 Approach
2.1 Global Matcher
2.2 Local Matcher
2.3 Match Validation
2.4 Global Optimization
3 Results
4 Conclusions and Future Works
References
Cluster ICP: Towards Sparse to Dense Registration
1 Introduction
2 Related Work
2.1 Feature-Based Approaches
2.2 Dense Approaches
3 Proposed Method
3.1 Surface Normal Segmentation
3.2 CICP Matching Pipeline
3.3 Optimization Framework
4 Results
4.1 Dense-Sparse Registration with CICP
4.2 Comparison with Existing Methods
4.3 Dense-to-Dense Data
5 Conclusion
References
Markerless Ad-Hoc Calibration of a Hyperspectral Camera and a 3D Laser Scanner
1 Introduction
2 Related Work
3 Generating Hyperspectral Panoramas
4 Extrinsic Calibration
4.1 Cylindric Camera Model
4.2 Mutual Information
4.3 Smoothing the Objective Function
5 Experiments
5.1 Qualitative Results
5.2 Run Time Analysis
5.3 Application Example
6 Conclusion
References
Predicting the Next Best View for 3D Mesh Refinement
1 Introduction
2 Related Works
3 Proposed Method
3.1 Photo-Consistent Reconstruction Index
3.2 Next Best View
4 Experimental Validation
5 Conclusions and Future Works
References
Robotic Applications
Grasping Strategies for Picking Items in an Online Shopping Warehouse
1 Introduction
1.1 Application
1.2 Summary
2 Related Work
3 Proposed Approach
3.1 HAF
3.2 AGILE
4 Experimentation and Results
4.1 Experimentation Procedures
4.2 Preliminary Object to Gripper Matching
4.3 Preliminary Implementation of Algorithms
4.4 Testing Objects in Isolation
5 Conclusion
References
Learning Based Industrial Bin-Picking Trained with Approximate Physics Simulator
1 Introduction
2 Related Works
3 Physics Simulator
4 Learning Based Approach
4.1 Convolutional Neural Network
4.2 Discriminator
4.3 Optimum Grasping Pose Detection
5 Collection of Training Data
6 Results
6.1 Discrimination
6.2 Derivation of Optimum Grasping Pose
6.3 Analysis of Model Approximation
7 Conclusions
References
Tool Exchangeable Grasp/Assembly Planner
1 Introduction
2 Definitions
3 Assembly/Grasp Planner
3.1 Placing Pose
3.2 Grasping Posture Set
3.3 Assembly Graph Search
3.4 Assembly Sequence
4 Results
5 Conclusion
References
Daily Assistive Robot Uses a Bag for Carrying Objects with Pre-contact Sensing Gripper
1 Introduction
1.1 Related Works
1.2 Paper Contribution
2 Daily Assistive Robot System for Carrying Objects with Bags
2.1 Overview
3 Pre-contact Manipulation for Daily Assistive Tasks with Bags
3.1 Finding Bag Handle
3.2 Opening a Bag
3.3 Putting the Robot Hand into the Bag
3.4 Object Grasping
4 Daily Assistive Robot Uses a Bag for Carrying Objects Experiments
5 Conclusion
References
Design of an Adaptive Force Controlled Robotic Polishing System Using Adaptive Fuzzy-PID
1 Introduction
2 Design and Setup of the Polishing Tool
3 System Dynamics
4 Fuzzy-PID Controller for Polishing
5 Experimental Setup and Results
5.1 Experimental Setup
5.2 Results and Discussion
6 Conclusions
References
Efficient, Collaborative Screw Assembly in a Shared Workspace
1 Introduction
1.1 Motivation
1.2 State of the Art
1.3 Problem Description
1.4 Goal
1.5 Concept
1.6 Structure
2 Approach
2.1 Hardware Setup
2.2 Challenges
2.3 Solution
2.4 Highlights
3 Results and Experiments
3.1 Experiments
3.2 Results
3.3 Discussion
4 Conclusions and Future Work
References
KittingBot: A Mobile Manipulation Robot for Collaborative Kitting in Automotive Logistics
1 Introduction
2 Related Work
3 System Overview
4 Perception
4.1 Container Detection
4.2 Part Segmentation
4.3 Parts Variant Recognition
5 Manipulation
6 Experiments
6.1 Showcase Setup
6.2 Kitting
6.3 Additional Experiments
7 Conclusion
References
User-Friendly Intuitive Teaching Tool for Easy and Efficient Robot Teaching in Human-Robot Collaboration
1 Introduction
2 Concept of Intuitive Teaching Device
2.1 Requirements
2.2 Design Concept
3 Jog Mechanism for Six DoF Motion Operation
4 Intuitive Teaching Method
5 Experiments
6 Conclusion and Future Research
References
A Critical Reflection on the Expectations About the Impact of Educational Robotics on Problem Solving Capability
1 Introduction
2 The Uncatchable Problem Solving Capability
2.1 The Three Components Model
2.2 A Short Review of Educational Robotics Applied to Problem Solving
3 Description of the Course
3.1 Setting and Participants
3.2 The Reasons Behind the Choice of Robot LEGO
3.3 Method and Activities
3.4 The Questionnaire
4 Data Analysis
4.1 The Course's Evaluation
4.2 The Perception of Robotics
4.3 The Attitude Towards Learning with Robotics
4.4 The Team Working Experience
4.5 The Way to Approach Problems
5 The Essential Step of Awareness
6 Considerations for the Future
7 Conclusions
References
Inferring Capabilities by Experimentation
1 Introduction
2 Related Work
3 Capability Models
4 Building Capability Models
4.1 Active Learning for Bayesian Networks
4.2 Model Refinement
5 Quantifying Capabilities
6 Results
6.1 BallKick Task
6.2 Pickup Task
7 Conclusion and Future Work
References
Triggering Robot Hand Reflexes with Human EMG Data Using Spiking Neurons
1 Introduction
2 Methods
2.1 Human EMG Data Interface and Training Data
2.2 Sub Network for EMG Classification
2.3 Sub Network for Motion Representation for Reflexes
2.4 Mapping to Robot Motor Commands
2.5 Integration Off All Components
3 Results
3.1 SNN Implementation
3.2 Training Data
3.3 Processing of EMG Data and Classification
3.4 Motion Generation
3.5 Interface to the Robot Hand
4 Discussion
References
A Localizability Constraint-Based Path Planning Method for Unmanned Aerial Vehicle
1 Introduction
2 Localizability Estimation
2.1 Static Localizability Matrix
2.2 Localizability Measure
3 Localizability Constraint-Based Path Planning Method
3.1 Objective Function
3.2 Constraints
3.3 Extraction of Local and Global Passable Regions
3.4 Optimization Method
4 Simulation and Discussion
4.1 Simulation Set-Up
4.2 Urban Canyon Environment
4.3 Wilderness Environment
4.4 Localization Error Analysis
5 Conclusion
References
Using IMU Sensor and EKF Algorithm in Attitude Control of a Quad-Rotor Helicopter
Abstract
1 Introduction
2 Modeling of Quad-Rotor
3 Attitude Estimation of Quad-Rotor Using EKF
3.1 System Configuration
3.2 Attitude Control of Quad-Rotor Using IMU
4 Experiment and Result
5 Conclusion
Acknowledgements
References
Unmanned Aerial Vehicles in Wireless Sensor Networks: Automated Sensor Deployment and Mobile Sink Nodes
1 Introduction
2 System Architecture
3 Individual Components
3.1 Sensor Node Hardware
3.2 Releasing Mechanism
3.3 Unmanned Aerial Vehicle
3.4 Optimal Route Calculation
4 System Validation
5 Conclusions
References
A Multi-layer Autonomous Vehicle and Simulation Validation Ecosystem Axis: ZalaZONE
Abstract
1 Introduction
2 Education and Research
3 Proving Ground
3.1 Dynamic Platform
3.2 Braking Platform
3.3 Smart City Zone
3.4 High and Low Speed Handling Courses
3.5 Motorway Section
3.6 Rural Roads
3.7 High-Speed Oval
3.8 Communication Network
3.9 Control Concept
4 Public Road and Urban Environment
4.1 Real Urban Test Environment
4.2 National and Cross-Border Test Loops
5 Sampling Vehicles with Crowdsourced Traffic Cloud
6 Conclusion
References
Towards Large Scale Urban Traffic Reference Data: Smart Infrastructure in the Test Area Autonomous Driving Baden-Württemberg
1 Introduction
2 Related Work
3 Concept
3.1 Local Intelligent Infrastructure
3.2 Perception Problem Factorization
4 Implementation
4.1 Camera Calibration
4.2 Object Detection
4.3 Multi Sensor Association and Fusion
4.4 Temporal Association and Multi-Target Tracking
4.5 Vehicle-To-X Communication
5 Evaluation
5.1 Hardware Setup
5.2 Object Recognition
5.3 Vehicle-to-X
6 Conclusion
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Intelligent Autonomous Systems 15

Description

More details

Other editions

Additional editions

Content

System requirements