Multi-Agent Pursuit-Evasion Game Based on Organizational Architecture

Abstract

Multi-agent coordination mechanisms are frequently used in pursuit-evasion games with the aim of enabling the coalitions of the pursuers and unifying their individual skills to deal with the complex tasks encountered. In this paper, we propose a coalition formation algorithm based on organizational principles and applied to the pursuit-evasion problem. In order to allow the alliances of the pursuers in different pursuit groups, we have used the concepts forming an organizational modeling framework known as YAMAM (Yet Another Multi Agent Model). Specifically, we have used the concepts Agent, Role, Task, and Skill, proposed in this model to develop a coalition formation algorithm to allow the optimal task sharing. To control the pursuers' path planning in the environment as well as their internal development during the pursuit, we have used a Reinforcement learning method (Q-learning). Computer simulations reflect the impact of the proposed techniques. ACM CCS (2012) Classification: Computing methodologies → Artificial intelligence → Distributed artificial intelligence → Multi-agent systems Theory of computation → Theory and algorithms for application domains → Algorithmic game theory and mechanism design → Convergence and learning in games

Multi-Agent Dynamic Leader-Follower Path Planning Applied to the Multi-Pursuer Multi-Evader Game

“…This fact negatively impacts the autonomy of the approach and its application in the real world. To overcome this limitation in this proposed approach, we have based on the MAS task coordination mechanism based on YAMAM organizational model proposed in [7].…”

“…In [7], the authors used Q-learning [24] in order to allow the pursuers to move in a decentralized way with the aim of obtaining the maximum payoffs detected in environment cells. Knowing that the payoffs are calculated in relation to the distance separating each environment cell from the cell containing the concerned evader.…”

“…Regarding the task sharing between the pursuers, in [7] the authors based on the concepts of Agent, Role, Task, and Skills forming the YAMAM organizational model to develop a pursuit groups access mechanism allowing an equitable and stable grouping of the pursuers during the pursuits. In [30], they proposed a coalition formation algorithm for the student agents selecting the courses proposed by the university in E-learning system.…”

“…In MPMEG, in addition to the path planning, a task-sharing method should also be provided to determine which pursuers have to pursue each evader. In this paper, we have based on the task-sharing method proposed in [7] to allow the pursuit groups' formation.…”

“…In the recent research activities, MAS organizational modeling frameworks [6] are enormously used to the coordinate the tasks in the PEG. Recently in [7], the authors have based on the different concepts forming the Yet Another Multi-Agent Model (YAMAM) [8] to create an efficient pursuit groups' access mechanism. Furthermore, supervised and unsupervised machine learning methods [9] are used in combination with organizational models to improve the tasks' coordination.…”