Advances in Practical Applications of Agents, Multi-Agent Systems, and Complex Systems Simulation. The PAAMS Collection : 20th International Conference, PAAMS 2022, L'Aquila, Italy, July 13–15, 2022, Proceedings / edited by Frank Dignum, Philippe Mathieu, Juan Manuel Corchado, Fernando De La Prieta.

Format:

Book

Contributor:

Dignum, Frank, editor.

Series:

Lecture Notes in Artificial Intelligence, 2945-9141 ; 13616

Language:

English

Subjects (All):

Artificial intelligence.

Data structures (Computer science).

Information theory.

Software engineering.

Computer engineering.

Computer networks.

Artificial Intelligence.

Data Structures and Information Theory.

Software Engineering.

Computer Engineering and Networks.

Computer Communication Networks.

Local Subjects:

Artificial Intelligence.

Data Structures and Information Theory.

Software Engineering.

Computer Engineering and Networks.

Computer Communication Networks.

Physical Description:

1 online resource (529 pages)

Edition:

1st ed. 2022.

Place of Publication:

Cham : Springer International Publishing : Imprint: Springer, 2022.

Summary:

This book constitutes the proceedings of the 20th International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2022, held in L'Aquila, Italy in July 2022. The 37 full papers in this book were reviewed and selected from 67 submissions. Another 10 demonstrations papers were selected from 11 submissions are presented here as short papers. The papers deal with the application and validation of agent-based models, methods, and technologies in a number of key applications areas, including: advanced models and learning, agent-based programming, decision-making, education and social interactions, formal and theoretic models, health and safety, mobility and the city, swarms and task allocation.

Contents:

Intro

Preface

Organization

Contents

Main Track

An Open MAS/IoT-Based Architecture for Large-Scale V2G/G2V

1 Introduction

2 Background and Related Work

3 System Architecture

4 Agent Interactions

4.1 Implemented Agent Strategies

5 Experimental Evaluation

5.1 Simulating Algorithms and Mechanisms

6 Conclusions and Future Work

References

.26em plus .1em minus .1emInvestigating Effects of Centralized Learning Decentralized Execution on Team Coordination in the Level Based Foraging Environment as a Sequential Social Dilemma

2 Background

2.1 Multi Agent Reinforcement Learning (MARL)

2.2 Sequential Social Dilemmas

2.3 Centralized Learning Decentralized Execution (CLDE)

3 Related Work

3.1 MARL Coordination in SSDs

3.2 Learning Algorithms

4 LBF as a SSD

5 Experimental Design

6 Results

7 Conclusion

Agent Based Digital Twin of Sorting Terminal to Improve Efficiency and Resiliency in Parcel Delivery

2 Problem Statement

2.1 State of the Art Analysis Techniques

3 Approach

3.1 Agent Based Realization

3.2 Simulation-Led Experimentation Aid

4 Illustrative Case Study

5 Conclusion

Fully Distributed Cartesian Genetic Programming

2 Distributed Cartesian Genetic Programming

3 Results

3.1 Regression

3.2 N-Parity

3.3 Classification

4 Conclusion

Data Synchronization in Distributed Simulation of Multi-Agent Systems

2 Data Synchronization in Distributed MAS Simulations

3 Synchronization Modes

3.1 Read and Write Operations

3.2 Data Synchronization Interface

3.3 Specification of Proposed Modes

3.4 Properties

4 Experiments

4.1 Experimental Settings

4.2 Results

References.

Co-Learning: Consensus-based Learning for Multi-Agent Systems

2 Co-Learning Algorithm

2.1 Consensus-based Multi-Agent Systems

2.2 Algorithm Description

3 Validation of Co-Learning Algorithm

3.1 Convergence Analysis

3.2 Network Efficiency

3.3 Effect of Network Size

4 Execution Using SPADE Agents

4.1 Co-Learning in SPADE

4.2 Execution Example

5 Conclusions

Multiagent Pickup and Delivery for Capacitated Agents

2 Related Work

3 Problem Description

4 Method

4.1 TPMT

4.2 TPMC

5 Evaluation

5.1 Case Studies

5.2 Experimental Setup

5.3 Results

6 Conclusion

Using Institutional Purposes to Enhance Openness of Multi-Agent Systems

2 Artificial Institutions and Purposes

3 Implementing a Multi-agent System with and Without the Purpose Model

3.1 Implementation Without Institutions and Purposes

3.2 Implementation with Institutions and Purposes

4 Discussion About both Implementations

5 Conclusions and Future Work

Developing BDI-Based Robotic Systems with ROS2

3 A Multi-agent Robotic RT-BDI Architecture

3.1 Development Tool-Kit

4 Demonstration of the MA-RT-BDI Architecture

5 Related Work

Combining Multiagent Reinforcement Learning and Search Method for Drone Delivery on a Non-grid Graph

2 Model

2.1 Problem Definition

2.2 Formulate Problem as Dec-MDP

3.1 Search Methods

3.2 Dynamic Programming Methods

4 Algorithm

4.1 The Search Module

4.2 The MARL Module

4.3 The Training Process

5.1 Evaluation Settings

5.2 Evaluation Results

6 Discussion

6.1 The MARL-SA Position in MAPF Solutions.

6.2 Learning Agent Selection in MARL-SA

Hierarchical Collaborative Hyper-Parameter Tuning

2 Methodology

2.1 Agent-Based Hyper-Parameter Tuning

2.2 Guided Randomized Agent-Based Tuning Algorithm

3 Results and Discussion

Towards the Combination of Model Checking and Runtime Verification on Multi-agent Systems

2 Preliminaries

2.1 Models for Multi-agent Systems

2.2 Syntax

2.3 Semantics

2.4 Runtime Verification and Monitors

2.5 Negative and Positive Sub-models

3 Our Procedure

4 Our Tool

4.1 Experiments

Explaining Semantic Reasoning Using Argumentation

2.1 Agent Oriented Programming Languages

2.2 Argumentation Schemes

2.3 OWL Ontologies

3 Scenario

4 Querying Ontologies

5 Translating SWRL Rules into Argumentation Schemes

5.1 Translating Arguments to Natural Language Explanations

6 Related Work and Conclusions

How to Solve a Classification Problem Using a Cooperative Tiling Multi-agent System?

2.1 Aggregation of Classifiers

2.2 Multi-agent Systems

3 Smapy

3.1 General Principle

3.2 Agents

3.3 Feedback

3.4 Non-cooperative Situations

4 Comparison of Linear Classifiers Alone with Context Learning Approach

4.1 Input Data

4.2 Experimental Protocol

5 Results

5.1 Classification Accuracy

5.2 Decision Boundaries

Multi-agent Learning of Numerical Methods for Hyperbolic PDEs with Factored Dec-MDP

3 Background

3.1 Factored Dec-MDP

3.2 Hyperbolic PDEs and WENO Scheme

4 Problem Formulation and Analysis.

4.1 Numerical Methods as Multi-agent Systems

4.2 Analysis of Different Reward Formulations

5 Experiment Results

5.1 Euler Equations and Training Setup

5.2 Results and Discussions

Multi-agent-based Structural Reconstruction of Dynamic Topologies for Urban Lighting

2 Real-World and Simulated Environment

3 Objective and Methods

3.1 Agent Modelling and Formalisation

3.2 Local Neighborhood Discrimination

3.3 Evaluation Metrics

4 Results and Discussion

Control Your Virtual Agent in its Daily-activities for Long Periods

3 Agent Model Description

3.1 Global Model Structure

3.2 Agent Internal State

3.3 Decision-Making Model

3.4 Task Execution Model

4 Results

Multi-Agent Task Allocation Techniques for Harvest Team Formation

3.1 Problem Description

3.2 Solution Fitness

3.3 GA Approach

3.4 Auction Approach

4.1 Data

4.2 Metrics

5.1 Trade-off Between Staff Time and Execution Time

5.2 Comparison to Alternative Approaches

6 Summary and Future Work

Study of Heterogeneous User Behavior in Crowd Evacuation in Presence of Wheelchair Users

2 Related Works

3 Our Proposed Agent-Based Panic Model Involving Wheelchair Users

3.1 Agent Attributes

3.2 Agent Model

3.3 Agent Behavior Logic

4 Experimental Setup

4.1 Agent Parameters

4.2 Environmental Parameters

4.3 Performance Metrics

5 Performance Evaluation and Results

5.1 Rate of Evacuation

5.2 Agent Attribute Distribution Versus Exit Time

5.3 Correlation Coefficient Trend

6 Conclusion and Future Works

Agent-Based Modelling and Simulation of Decision-Making in Flying Ad-Hoc Networks

2 Motivation

2.1 System Overview

2.2 ConOps Agent Algorithm

3 Agent-Based Modelling and Simulation

3.1 Implementation and Usage

3.2 Evaluation Method

3.3 Evaluation Results

Deep RL Reward Function Design for Lane-Free Autonomous Driving

2.1 Deep Deterministic Policy Gradient

2.2 Related Work

3 Our Approach

3.1 The Lane-Free Traffic Environment

3.2 State Representation

3.3 Action Space

3.4 Reward Function Design

4 Experimental Evaluation

4.1 RL Algorithm and Simulation Setup

4.2 Results and Analysis

An Emotion-Inspired Anomaly Detection Approach for Cyber-Physical Systems Resilience

2.1 Resilience

2.2 Anomaly Detection for CPS Resilience

3 The Proposed Approach

4.1 System Description

4.2 Measures of Resilience

4.3 Scenario Description

4.4 Results and Evaluation

5 Conclusions and Perspectives

Bundle Allocation with Conflicting Preferences Represented as Weighted Directed Acyclic Graphs

2 Problem Model

3 Path Allocation Schemes

3.1 Utilitarian Allocation (util)

3.2 Leximin Allocation (lex)

3.3 Approximated Leximin Allocation (a-lex)

3.4 Greedy Allocation (greedy)

3.5 Round-Robin Allocations (p-rr and n-rr)

Shifting Reward Assignment for Learning Coordinated Behavior in Time-Limited Ordered Tasks

3.1 Agent and Environment

3.2 Task Structure

3.3 Time Limitation for Completing Each Task

4 Proposed Method.

4.1 Shifting Two-Stage Reward Assignment.

Notes:

Includes bibliographical references and index.

ISBN:

3-031-18192-1