Distributed Autonomous Robotic Systems

Name: Distributed Autonomous Robotic Systems | 15th International Symposium
Brand: Springer
Price: 255.73 EUR
Availability: OnlineOnly

15th International Symposium

Fumitoshi Matsuno Shun-Ichi Azuma Masahito Yamamoto(Editor)

Springer (Publisher)

1st Edition

Published on 3. January 2022

XIII, 444 pages

E-Book

PDF with digital watermarking

System requirements

978-3-030-92790-5 (ISBN)

€255.73incl. 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
Executive Committee
General Chair
Program Co-chairs
Vice Program Co-chairs
Publication Chairs
Finance Chair
Publicity Chair
Registration Chairs
Local Arrangement Chairs
Secretary
Advisory Committee
Contents
Generating Goal Configurations for Scalable Shape Formation in Robotic Swarms
1 Introduction
2 Preliminaries
2.1 Goal Configuration Generation: Problem Statement
2.2 Notations
3 Approach
3.1 Scaling
3.2 Interpolation
4 Performance Evaluation
4.1 Experiments on Generated Configurations
4.2 Experiments on Simulated Robotic Swarm
4.3 Experiments on Physical Robotic Swarm
5 Conclusion
References
Leading a Swarm with Signals
1 Introduction
2 Related Work
3 Leading with Signals in the Flocking Problem
3.1 Basic Model Empirical Evaluation
3.2 Swarm with Limited Sensing Range
3.3 Limited Signal
3.4 Leader Position and Multiple Leaders
4 Signaling in the Dispersion Problem
4.1 Dispersion Problem Definition
4.2 Dispersion Model Results
5 Conclusion and Future Work
References
Byzantine Fault Tolerant Consensus for Lifelong and Online Multi-robot Pickup and Delivery
1 Introduction
2 Background
2.1 MAPF and MAPD
2.2 Blockchain and Byzantine Faults
2.3 Tendermint
3 Methods
4 Benchmark Algorithms
5 Experimental Evaluation
5.1 Makespan and Service Time Comparison
5.2 Runtime Comparison
5.3 Byzantine Failure Comparison
6 Conclusion
References
Decentralized Multi-robot Planning in Dynamic 3D Workspaces
1 Introduction
2 Problem Formulation
2.1 Notation and Assumptions
2.2 Problem Statement
3 Approach
3.1 Workspace Representation
3.2 Kinodynamic Planner
4 Evaluation
4.1 Implementation Details and Experiment Design
4.2 Results
5 Conclusion and Future Work
References
Optimized Direction Assignment in Roadmaps for Multi-AGV Systems Based on Transportation Flows
1 Introduction
2 Scenario and Problem Definition
2.1 Scenario
2.2 Representation of the Roadmap
2.3 Representation of the Flows
2.4 Problem Definition
3 Proposed Method
4 Evaluation
5 Conclusions
References
Datom: A Deformable Modular Robot for Building Self-reconfigurable Programmable Matter
1 Introduction
2 Related Works
3 The Datom Model
3.1 Theoretical Geometry of the Deformable Module
3.2 Deformation Process of a Single Datom
3.3 Design of a Thick Datom
4 Motion Capabilities in an Ensemble
5 Simulation
5.1 Algorithms
5.2 Results
6 Conclusion
References
The Impact of Network Connectivity on Collective Learning
1 Introduction
2 Related Work
3 A Propositional Model for Collective Learning
4 Small-World Networks
5 Agent-Based Simulations
5.1 Convergence Results for Regular Small-World Networks
5.2 Convergence Results for Small-World Networks with Random Rewiring
6 Discussion and Conclusion
References
On the Communication Requirements of Decentralized Connectivity Control
1 Introduction
2 Related Work
3 Control Law
3.1 Connectivity Maintenance Contribution
3.2 Robustness Improvement Contribution
3.3 Coverage Improvement Contribution
4 Field Experiments
4.1 Robotic and Computing Hardware
4.2 Middleware and Software Implementation
4.3 Inter-robot Communication with Buzz
5 Results and Discussion
5.1 Timing Performance
5.2 Connectivity
5.3 Robustness
5.4 Coverage
6 Conclusions
References
Behavioral Simulations of Lattice Modular Robots with VisibleSim
1 Introduction
2 Related Works
3 Simulator Overview
4 Programming Environment and Features
4.1 User Application Demonstration
4.2 User Interactions
4.3 Customization Hooks
4.4 Export Tools
5 Usage and Evaluation
5.1 Use Cases
5.2 Simulation Fidelity
5.3 Scalability
6 Conclusion and Future Work
References
Evolving Robust Supervisors for Robot Swarms in Uncertain Complex Environments
1 Introduction
2 Related Work
3 Methodology
3.1 Simulation
3.2 Exploring Uncertain Environment States
3.3 Swarm Supervisor
3.4 Evolving the Swarm Supervisor
4 Results
4.1 Behaviour Benchmarks
4.2 Evolved Supervisory Control
4.3 Qualitative Analysis of Evolved Supervision
4.4 Robust Search Strategies
4.5 Performance Variation Between Environments
5 Conclusions and Future Work
References
Distributed Cooperative Localization with Efficient Pairwise Range Measurements
1 Introduction
2 Methodology
2.1 State Description
2.2 Motion and Individual Measurements
2.3 Range-Based Covariance Intersection
2.4 Peer Selection
2.5 Range Queries and Communication
3 Experimental Setup
3.1 Simulation
3.2 Outdoor Experiments
4 Results
4.1 Evaluation Metrics
4.2 Simulations
4.3 Outdoor Experiments
4.4 Computation and Data Overhead
5 Conclusion
References
Robust Localization for Multi-robot Formations: An Experimental Evaluation of an Extended GM-PHD Filter
1 Introduction
2 Background
2.1 The FI-GM-PHD Filter
2.2 Graph-Based Formation Control
2.3 Role Assignment
3 Experimental Campaign
3.1 Implementation
3.2 Self-localization and Measurement Errors
3.3 Tailoring the Probability of Detection
3.4 Evaluation Metrics
3.5 Scenarios
4 Results
4.1 Scenario I: Multi-robot Tracking
4.2 Scenario II: Tracking for Formation Control
4.3 Scenario III: Realistic Environment
5 Conclusion
References
Opportunistic Multi-robot Environmental Sampling via Decentralized Markov Decision Processes
1 Introduction
2 Related Work
3 Problem Setup and Basic Algorithm
3.1 Prediction via Gaussian Processes (GPs)
3.2 Control via Markov Decision Processes (MDPs)
4 Algorithm with Opportunistic Communication
4.1 Prediction via GP Mixtures
4.2 Control via Decentralized MDPs
5 Evaluation
6 Conclusions and Future Work
References
A PHD Filter Based Localization System for Robotic Swarms
1 Introduction
2 Problem Setting
3 Multi-sensor PHD Filter
4 PHD-Filter Based Relative Localization Module
4.1 Time Update
4.2 Lidar Measurement Update
4.3 Camera Measurement Update
5 Experiments
5.1 Results
6 Conclusions
References
An Innate Motivation to Tidy Your Room: Online Onboard Evolution of Manipulation Behaviors in a Robot Swarm
1 Introduction
2 Experiment Design
2.1 Robots in Simulation and Hardware
2.2 Environment
2.3 Evolution of Robot Controllers
2.4 Hardware Protection
3 Results
3.1 Simulation
3.2 Experiments on Physical Robots
4 Conclusion
References
Multi-agent Reinforcement Learning and Individuality Analysis for Cooperative Transportation with Obstacle Removal
1 Introduction
2 Problem Statement
3 Policy Optimization
4 Numerical Experiments
4.1 Validation Test
4.2 Results
4.3 Individuality Analysis
5 Demonstration
6 Conclusion and Future Work
References
Battery Variability Management for Swarms
1 Introduction
2 Related Work
3 System Architecture
4 The Swap Algorithm
5 Field Experiments
5.1 Battery Life Analysis
5.2 Mission Performance
6 Simulation Experiments
6.1 Field Exercise Mission Plans with and Without Swap
6.2 Replacement Pool Size
6.3 Swap Threshold
7 Discussion
8 Conclusion
References
Spectral-Based Distributed Ergodic Coverage for Heterogeneous Multi-agent Search
1 Introduction
2 Prior Work
2.1 Multi-agent Search
2.2 Heterogeneity in MAS
3 Background on Ergodic Search Processes
4 Distributed Heteregeneous Ergodic Search
4.1 Spectral Bands of the Information Distribution
4.2 Assignment of Agents to Spectral Bands
5 Results and Discussion
5.1 Agent's Sensing and Motion Models
5.2 Experiment Details
5.3 Experimental Results
6 Conclusion
References
Multi-agent Deception in Attack-Defense Stochastic Game
1 Introduction
2 Related Work
3 Technical Approach
3.1 Problem Formulation
3.2 Nash Equilibrium and Bayesian Nash Equilibrium
3.3 Attacker Strategy and Belief Update
3.4 Deceptive Planning
3.5 Game Tree Sampling
4 Results
5 Conclusion and Future Research
References
Tractable Planning for Coordinated Story Capture: Sequential Stochastic Decoupling
1 Motivation
2 Related Work
3 Definitions and Problem Statement
3.1 Worlds and Narratives: Event Model and Story Automaton
3.2 Robot Model
3.3 Policies and Problem Statement
4 Solving Mrrcm
4.1 Preliminary Definitions
4.2 Full Joint Plan
4.3 Sequentialized Planning
5 Case Study
6 Conclusion
References
Errors in Collective Robotic Construction
1 Introduction
2 Terminology
3 Influence of Policy Choice on System Performance
4 Error Analysis
5 Mitigating Errors Through Predictive Local Checks
5.1 Predictive Local Checks
5.2 Impact of Predictive Local Checks on Construction Time
6 Conclusion
References
Optimal Multi-robot Perimeter Defense Using Flow Networks
1 Introduction
2 Related Work
3 Problem Formulation
4 Flow Network Formulation
4.1 Graph Generation
4.2 Intuition Behind Constructed Graph
4.3 Recovery of Intruder Assignments
4.4 Optimality of Intruder Assignments
4.5 Complexity
5 Extension to Heterogeneous Defenders
5.1 Deconflict Assignment
5.2 Intuition Behind Deconflict Assignment
5.3 Complexity
6 Results
6.1 Experimental Setup
6.2 Discussion of Results
7 Conclusion
References
Classification-Aware Path Planning of Network of Robots
1 Introduction
2 Related Work
3 Distributed Classification Architecture
3.1 Preliminaries and Feature Extraction
3.2 LSTM Feature Encoding
3.3 Classification-Aware Path Planning
3.4 Map Classifier and Global Classification Reward
4 Training Preliminaries
5 Case Studies
5.1 Simulation with the Campus Map Dataset
5.2 Efficient Distributed Classification
5.3 The Coverage of Informative Regions
6 Conclusions
References
Monitoring and Mapping of Crop Fields with UAV Swarms Based on Information Gain
1 Introduction
2 Problem Description
2.1 World Model and UAV Swarm Simulation
2.2 Model of Weed-Classification Uncertainty
2.3 Uncertainty Reduction from Multiple Observations
3 Reinforced Random Walks for Monitoring and Mapping
3.1 Neighbourhood Selection Strategy
3.2 Baseline Strategy Based on Optimal Pre-planned Trajectories
3.3 Baseline Strategy Based on Potential Fields
4 Experimental Results
5 Conclusions
References
A Discrete Model of Collective Marching on Rings
1 Introduction
2 Model and Definitions
3 Stabilization Analysis
3.1 Locusts on Narrow Ringlike Arenas (k=1)
3.2 Locusts on Wide Ringlike Arenas (k & 1)
4 Simulation and Empirical Evaluation
5 Concluding Remarks
References
Map Learning via Adaptive Region-Based Sampling in Multi-robot Systems
1 Introduction
2 Related Work
3 Problem Description: Multi-robot IPP for Map Learning Under Time and Communication Limitations (IPP-MRS)
4 Tackling IPP-MRS with Different Architectures
5 Fully Distributed Adaptive Selection of ROI and Paths
5.1 Identification of the ROI with the Largest Utility
5.2 Defining Predicted Utility Values
5.3 Identify Candidate Sampling Locations Inside Selected ROI
5.4 Compute a Resourceful and Informative Path Using Orienteering
5.5 Executing the Plan, Termination and Replanning
6 Leader-Follower Architecture
7 Computational Experiments
7.1 Accuracy in Map Estimation
7.2 Scalability in Performance and in Computations
8 Conclusions
References
Collective Transport via Sequential Caging
1 Introduction
2 Related Work
3 Methodology
3.1 Problem Formulation
3.2 Task Allocation Based Caging
3.3 Behaviors for Pushing and Rotating
4 Experiments
5 Conclusions
References
ReactiveBuild: Environment-Adaptive Self-Assembly of Amorphous Structures
1 Introduction
2 Algorithm Description
3 Simulator Overview
4 Algorithm Validation
4.1 Tower
4.2 Chain
4.3 Cantilever
4.4 Bridge
5 Conclusion
References
Processes for a Colony Solving the Best-of-N Problem Using a Bipartite Graph Representation
1 Introduction
2 Related Literature
3 Bipartite Graph Formalism
4 Attachment and Detachment Probabilities
4.1 Attachment Probability
4.2 Detachment Probability
5 Methods
6 Results
7 Conclusion and Future Work
References
Decentralized Navigation in 3D Space of a Robotic Swarm with Heterogeneous Abilities
1 Introduction
2 Problem Setting
2.1 Modeling of Agents
2.2 Connectivity of Agents
2.3 Control Objective
3 Proposed Method
3.1 Leader's Speed Constraint
3.2 Target Selection Process
3.3 Control Input
3.4 What Is Guaranteed by the Proposed Method
4 Mathematical Proof
5 Simulation
6 Conclusion
References
Using Reinforcement Learning to Herd a Robotic Swarm to a Target Distribution
1 Introduction
2 Methodology
2.1 Problem Statement
2.2 Design of Leader Control Policies Using Temporal-Difference Methods
3 Simulation Results
4 Experimental Results
5 Conclusion and Future Work
References
Preservation of Giant Component Size After Robot Failure for Robustness of Multi-robot Network
1 Introduction
2 Definition and Formulation
3 Modification for Perturbed Algebraic Connectivity
4 Threshold Computation
5 Component Size Preservation Control
5.1 Simulation Examples
6 Conclusion
References
Swarm Localization Through Cooperative Landmark Identification
1 Introduction
2 Problem Setting
3 Methodology
3.1 System Architecture
3.2 Cooperative Landmark Identification
3.3 Particle Filter for Localization
4 Simulations
5 Conclusions
References
Author Index

System requirements

Save as PDF Copy link into clipboard

Schweitzer Fachinformationen

Distributed Autonomous Robotic Systems

Description

More details

Other editions

Additional editions

Content

System requirements