Distributed Autonomous Robotic Systems: 16th International Symposium

Foreword
Preface
Contents
How Can We Understand Multi-Robot Systems? a User Study to Compare Implicit and Explicit Communication Modalities
	1 Introduction
	2 Methods
		2.1 Communication Modalities
		2.2 Experimental Task
		2.3 Experimental Setup
		2.4 Participants
		2.5 Data Collection and Analysis
	3 Results
		3.1 Correctness of Responses
		3.2 Response Time
		3.3 Subjective Reporting
	4 Conclusion
	References
The Benefits of Interaction Constraints in Distributed Autonomous Systems
	1 Introduction
	2 Collective Learning in Multi-agent Systems
	3 A Three-Valued Model for Collective Learning
	4 Simulation Environment
	5 The Networks Underlying Agent Interactions
	6 Multi-agent Simulation Experiments
		6.1 Convergence Results for Physical Networks
	7 Constraining the Interaction Network
		7.1 Convergence Results for Constrained Interaction Networks
	8 Conclusion and Future Work
	References
Outlining the Design Space of eXplainable Swarm (xSwarm): Experts\' Perspective
	1 Introduction
	2 Background and Related Work
	3 Background and Related Work
		3.1 Participants
		3.2 Study Procedure and Design
		3.3 Analysis
	4 Results
		4.1 Explanation Categories
		4.2 Challenges Faced by Swarm Experts
	5 Conclusion
	References
VMAS: A Vectorized Multi-agent Simulator for Collective Robot Learning
	1 Introduction
	2 Related Work
	3 The VMAS Platform
	4 Multi-robot Scenarios
	5 Comparison with MPE
	6 Experiments and Benchmarks
	7 Conclusion
	References
FLAM: Fault Localization and Mapping
	1 Introduction
	2 Related Work and Background
		2.1 Fault Detection
		2.2 Risk Awareness
		2.3 Distributed Storage
	3 System Model
		3.1 Risk Modelling
		3.2 Feature Vectors
		3.3 Fault Detection
		3.4 Distributed Belief Map
		3.5 Implementation
	4 Simulations
		4.1 Experimental Setup
		4.2 Results
	5 Conclusion
	References
Social Exploration in Robot Swarms
	1 Introduction
	2 Related Work
	3 Methodology
		3.1 Simulation and Local Swarm Behaviours
		3.2 Generating Random Environments
		3.3 Agents with ``Happiness\'\'
		3.4 Social Exploration Algorithm
	4 Results
		4.1 Performance in Randomly Generated Environments
	5 Conclusions and Future Work
	References
Stochastic Nonlinear Ensemble Modeling and Control for Robot Team Environmental Monitoring
	1 Introduction
	2 Problem Formulation
		2.1 Task Topology
		2.2 Ensemble Model
	3 Ensemble Model Control
		3.1 Feedback Controller
		3.2 Distributed Microscopic Algorithm
	4 Simulation and Experimental Setup
	5 Results
	6 Discussion
	7 Conclusion
	References
A Decentralized Cooperative Approach to Gentle Human Transportation with Mobile Robots Based on Tactile Feedback
	1 Introduction
	2 Methodology
		2.1 Flexible Tactile Sensors and Methods of Acquiring Force Information
		2.2 Design of Control Law
	3 Modeling and Simulation
		3.1 Modeling
		3.2 Simulation
	4 Experiments
		4.1 Experiment Setup
		4.2 Experiment Result
	5 Conclusion and Future Work
	References
Sparse Sensing in Ergodic Optimization
	1 Introduction
	2 Prior Work
	3 Sparse Sensing in Ergodic Optimization
		3.1 Sparse Ergodic Optimization
		3.2 Multi-agent Sparse Ergodic Optimization
	4 Results and Discussion
		4.1 Experimental Details
		4.2 Single Agent Sensing Distribution
		4.3 Multi-agent Sparse Ergodic Optimization
	5 Conclusion
	References
Distributed Multi-robot Tracking of Unknown Clustered Targets with Noisy Measurements
	1 Introduction
	2 Problem Formulation
		2.1 Lloyd\'s Algorithm
	3 Distributed Multi-target Tracking
		3.1 Instantaneous State Estimation
		3.2 Cumulative State Estimation
		3.3 Environment Approximation
		3.4 Distributed Control
		3.5 Algorithm Outline
	4 Simulations
		4.1 Qualitative Comparison
		4.2 Quantitative Comparison
	5 Conclusions
	References
A Force-Mediated Controller for Cooperative Object Manipulation with Independent Autonomous Robots
	1 Background
	2 Related Work
	3 Challenges and Assumptions
	4 Control Methodology
		4.1 Closed Loop Control Dynamics
		4.2 Inertial Estimation and Error Compensation
		4.3 Controller Stability
		4.4 Task-Frame Switching
	5 Experimental Validation
		5.1 Implementation in Hardware
		5.2 Implementation in Simulation
	6 Results
		6.1 Controller Performance Within Static Task Domain
		6.2 Controller Performance with a Changing Task Domain
		6.3 Multi-robot Performance with Changing Task Domain
	7 Discussion
	References
A Distributed Architecture for Onboard Tightly-Coupled Estimation and Predictive Control of Micro Aerial Vehicle Formations
	1 Introduction
	2 Problem Formulation
	3 Methodology
		3.1 Follower Estimation Scheme
		3.2 Follower Controller Scheme
	4 Experiments and Results
		4.1 Formation Navigation
		4.2 Formation Reconfiguration
		4.3 Discussion
	5 Conclusion
	References
Search Space Illumination of Robot Swarm Parameters for Trustworthy Interaction
	1 Introduction
	2 Methodology
		2.1 Scenario
		2.2 Swarm Metrics
		2.3 Map-Elites Illumination Algorithm Methodology
	3 Results
		3.1 Qualitative Analysis: Case Study
	4 Conclusion and Future Work
	References
Collective Gradient Following with Sensory Heterogeneous UAV Swarm
	1 Introduction
	2 Methodology
	3 Experimental Setup
	4 Results and Discussion
	5 Conclusion
	References
DAN: Decentralized Attention-Based Neural Network for the MinMax Multiple Traveling Salesman Problem
	1 Introduction
	2 Prior Works
	3 Problem Formulation
	4 mTSP as an RL PROBLEM
	5 DAN: Decentralized Attention-Based Network
	6 Training
	7 Experiments
		7.1 Results
		7.2 Discussion
	8 Conclusion
	References
Receding Horizon Control on the Broadcast of Information in Stochastic Networks
	1 Introduction
	2 Background and Problem Formulation
		2.1 Graph Theory
		2.2 Information Propagation Model
	3 Methodology
		3.1 Control Strategy
		3.2 Expected Time to Broadcast Information
		3.3 Robust Moment Closure
	4 Numerical Evaluation and Discussions
	5 Conclusions and Future Work
	References
Adaptation Strategy for a Distributed Autonomous UAV Formation in Case of Aircraft Loss
	1 Introduction
	2 Adaptation Algorithm for a Decentralized UAV Formation
		2.1 Distributed Autonomous System Model
		2.2 Adaptation Strategy for a Distributed Autonomous System
	3 Simulation Results and Discussion
	4 Conclusions
	References
DGORL: Distributed Graph Optimization Based Relative Localization of Multi-robot Systems
	1 Introduction
	2 Related Work
	3 Problem Formulation and the Proposed DGORL Solution
		3.1 Graph Formation
		3.2 Expansion Through Transition
		3.3 Optimization
	4 Theoretical Analysis
	5 Simulation Experiments and Results
	6 Conclusion
	References
Characterization of the Design Space of Collective Braitenberg Vehicles
	1 Introduction
		1.1 Related Work
	2 Model and Simulation
		2.1 Agent Model
		2.2 World
		2.3 Metrics
	3 Emergent Behaviors
		3.1 Behavior Characterization
		3.2 General Observations
		3.3 Love
		3.4 Aggression
		3.5 Fear
		3.6 Curiosity
	4 Conclusion and Outlook
	References
Decision-Making Among Bounded Rational Agents
	1 Introduction
	2 Related Work
	3 Problem Formulation
		3.1 Multi-agent Markov Decision Process
		3.2 Iterative Best Response for MMDPs
	4 Methodology
		4.1 Bounded Rational Agents in Game-Theoretic Framework
		4.2 Importance Sampling for Computing Bounded-Rational Strategies
	5 Simulated Experiments
		5.1 Simulation Setup
		5.2 Results
	6 Physical Experiments
	7 Conclusion
	References
Distributed Multi-robot Information Gathering Using Path-Based Sensors in Entropy-Weighted Voronoi Regions
	1 Introduction
	2 Problem Formulation
	3 Distributed Information Gathering
		3.1 Distributed Entropy Voronoi Partition and Planner (DEVPP)
		3.2 Multi-agent Distributed Information-Theoretic Planner (MA-DITP)
		3.3 Multi-agent Global Information-Theoretic Planner (MA-GITP)
	4 Experiments and Results
	5 Conclusion
	References
Distributed Multiple Hypothesis Tracker for Mobile Sensor Networks
	1 Introduction
	2 Background
		2.1 MHT Definition
		2.2 Lloyd\'s Algorithm
	3 Distributed Multiple Hypothesis Tracker
		3.1 Assumptions
		3.2 Track Exchange and Fusion
		3.3 Importance Weighting Function
	4 Results
		4.1 Performance Metric
		4.2 Stationary Targets
		4.3 Dynamic Targets
		4.4 Discussion
	5 Conclusion
	References
Distributed Multirobot Control for Non-cooperative Herding
	1 Introduction
	2 Prior Work
	3 Problem Formulation
	4 Controller Design
		4.1 Approach 1: One Dog to One Sheep Allocation Based Approach
		4.2 Approach 2: Iterative Distributed Reformulation of (12)
	5 Results
		5.1 Simulation Results
		5.2 Robot Experiments
	6 Conclusions
	7  Appendix: Proof of Feasibility for Approach 1
	References
On Limited-Range Coverage Control for Large-Scale Teams of Aerial Drones: Deployment and Study
	1 Introduction
	2 Methods
		2.1 Coverage Control Strategy
		2.2 Simulation Protocol
		2.3 Field Test Protocol
		2.4 Metrics
	3 Results and Discussion
		3.1 Simulation Results
		3.2 Results of Field Tests
	4 Conclusions
	References
MARLAS: Multi Agent Reinforcement Learning for Cooperated Adaptive Sampling
	1 Introduction
	2 Problem Formulation
	3 Approach
	4 Experiments and Results
		4.1 Analyzing Scalability of Learned Policy
		4.2 Analyzing Robustness of the Learned Policy
		4.3 Online Adaptations to the Changing Field
		4.4 Robot Experiments
	5 Conclusion and Future Work
	References
Proportional Control for Stochastic Regulation on Allocation of Multi-robots
	1 Introduction
	2 Problem Formulation
		2.1 Graph Theory
		2.2 Ensemble Model
	3 Methodology
		3.1 Control Strategy
		3.2 Control Analysis
	4 Results
		4.1 Numerical Simulations
		4.2 Experimental Results
	5 Discussion
	6 Conclusions and Future Work
	References
Comparing Stochastic Optimization Methods for Multi-robot, Multi-target Tracking
	1 Introduction
		1.1 Related Work
		1.2 Contributions
	2 Problem Formulation
		2.1 PHD Filter
		2.2 Lloyd\'s Algorithm
		2.3 Distributed PHD Filter
		2.4 Assumptions
	3 Distributed Control
		3.1 Simulated Annealing
		3.2 Particle Swarm Optimization
		3.3 Communication and Complexity
	4 Experimental Results
		4.1 Evaluation Metric
		4.2 Stationary Targets
		4.3 Dynamic Targets
	5 Conclusion
	References
Multi-agent Deep Reinforcement Learning for Countering Uncrewed Aerial Systems
	1 Introduction
		1.1 Related Work
	2 Problem Formulation
	3 Multi-agent Deep Reinforcement Learning for C-UAS
		3.1 Proximal Policy Optimization Based Advantage Actor Critic
		3.2 Parameter Sharing
	4 Simulation Results
	5 Conclusion
	References
Decentralized Risk-Aware Tracking of Multiple Targets
	1 Introduction
	2 Related Work
	3 Problem Formulation
		3.1 Notation
		3.2 Robot and Target Modeling
		3.3 Problem Definition
	4 Approach
		4.1 Decentralized Goal Position Generation
		4.2 Decentralized Motion Synthesis
		4.3 Decentralized Target Position Estimation
	5 Experiments
		5.1 Decentralized Tracking
		5.2 Comparison to Centralized Target Tracking
	6 Conclusion and Future Work
	References
Application of 3D Printed Vacuum-Actuated Module with Multi-soft Material to Support Handwork Rehabilitation
	1 Introduction
	2 MORI-A FleX: The Vacuum-Actuated Module with 3D Printed Flexible Connectors
		2.1 Creation Procedure of MORI-A FleX
		2.2 TPU Internal Construction
		2.3 3D Printed Flexible Connector
	3 MORI-A Flex Mechanical Properties
		3.1 Load Displacement Performance
		3.2 Coefficient of Friction with Texture Change Connector
	4 Applications
	5 Conclusion
	References
Combining Hierarchical MILP-MPC and Artificial Potential Fields for Multi-robot Priority-Based Sanitization of Railway Stations
	1 Introduction and Related Works
	2 Architecture
	3 Experiments and Comparison with MARL Technique
	4 Conclusions
	References
Exploration System for Distributed Swarm Robots Using Probabilistic Action Decisions
	1 Introduction
	2 The Proposed Exploration Swarm Robot System and Area Control Algorithm
		2.1 Distributed Robot System for Wide-Area Exploration by Swarm Robots
		2.2 Overview of the Region Control Algorithm with Probabilistic Action Decision
		2.3 Algorithm for Probabilistic Action Decision Making
	3 Evaluation by Simulation
		3.1 Algorithm Verification by Simulator and Comparison of Search Efficiency by Changing Control Parameters
	4 Development and Implementation of Exploration Systems
		4.1 Hardware Development of the Robot for Actual Equipment Verification
		4.2 Software and Exploration System Configuration
	5 Exploration Field Tests and Results
		5.1 Description of Experimental Environment/conditions and Data Collected
		5.2 Examination of Collected Data
	6 Conclusion
	References
Distributed Estimation of Scalar Fields with Implicit Coordination
	1 Introduction
	2 Related Work
	3 Problem Statement
	4 Methods
		4.1 Gaussian Process Regression
		4.2 Spatial Prior
		4.3 Exploration
		4.4 Vertex Quality Computation
	5 Experimental Evaluation and Discussion
	6 Conclusion
	References
A Constrained-Optimization Approach to the Execution of Prioritized Stacks of Learned Multi-robot Tasks
	1 Introduction
	2 Background and Related Work
		2.1 Multi-task Learning, Composition, and Execution
		2.2 Constraint-Based Task Execution
		2.3 From Dynamic Programming to Constraint-Driven Control
	3 Prioritized Multi-task Execution
	4 Experimental Results
		4.1 Multi-robot Tasks
		4.2 Discussion
	5 Conclusion
	A Comparison Between Optimal Control, Optimization-Based Control, and RL Policy
	B Implementation Details
	References
A Comparison of Two Decoupled Methods for Simultaneous Multiple Robots Path Planning
	1 Introduction
	2 Related Work
	3 Planning Algorithms
		3.1 Problem Formulation
		3.2 Decoupled Prioritized Planning (PP)
		3.3 Decoupled Fixed-Path Coordination (FPC)
	4 Numerical Experiments
		4.1 Illustrative Test Cases
		4.2 Monte Carlo Simulations
	5 Conclusion
	References
Smarticle 2.0: Design of Scalable, Entangled Smart Matter
	1 Introduction
	2 Background and Related Work
	3 Electro-mechanical Design
	4 Fabrication
	5 Characterization
	6 Multi-robot Demonstrations
	7 Summary
	References
Hybrid Flock - Formation Control Algorithms
	1 Introduction
	2 Methodology
		2.1 Pot2Lapl
		2.2 Flock2Lapl
	3 Experimental Evaluation
		3.1 Robotic Platform
		3.2 Evaluation Metrics
		3.3 Experimental Setup
		3.4 Characterization Experiment
		3.5 Narrow Passage Experiment
	4 Results and Discussion
	5 Conclusion
	References
Multi-Robot Systems Research: A Data-Driven Trend Analysis
	1 Introduction
	2 Exhaustive Search
		2.1 Methodology
		2.2 Findings from Journal-Specific Analysis
		2.3 Findings from Trend Analysis
	3 Expert Survey
		3.1 Methodology
		3.2 Findings
	4 Conclusions
	References
Coverage Control for Exploration of Unknown Non-convex Environments with Limited Range Multi-robot Systems
	1 Introduction
	2 Problem Statement
	3 Distributed Control Algorithm for Exploration
	4 Time-Varying Probability Density Function for Exploration
	5 Simulations and Experiments
		5.1 Implementation
		5.2 Results and Discussion
	6 Conclusions
	References
Author Index