COOPERATIVE DELIVERY MODEL

Cooperative Delivery Model

Cooperative Delivery Model

Introduction

Being motivated by earlier results and inspired by the properties of self-regulation in biological immune systems, this paper develops a cooperative control model that is based on the behavioral model to provide an effective methodology to coordinate and control multiagent systems. The core of the control model exploits the immunity-based regulation mechanisms to emerge group collective behaviors and cooperative strategies in achieving global goals. Through the exposition of the immune system characteristics, agents in the framework are abstracted as independent entities, known as AIS agents, which are operating in a multiagent system that carry local information, search for solutions and exhibit robust behavior to accomplish tasks. (Tsumura 2004)

Discussion

Within the scope of the proposed control model, the adopted immunity-based regulation mechanism comprised three important threshold measures that regulate the activities of AIS agents and provide effective coordination for multiagent systems under three domains, namely, exploration, achievement and cooperation domains. These three domains represent a typical operation cycle of AIS agents in performing activities such as exploration, collaboration, tackling tasks, etc. In each of these domains, a threshold measure is introduced such that AIS agents act and behave accordingly under different situations encountered. These threshold measures being the key elements for effective agent coordination form the heart of the immunity-based regulation mechanism. (Singh & Thayer 2001)

Specifically, the three threshold measures that govern the regulation mechanisms are: (i) wandering threshold; (ii) affinity threshold and (iii) activation threshold. The wandering threshold is employed under the exploration domain. In particular, the number of AIS agents that perform exploration activities are regulated within a wandering zone. The agents are regulated to perform either random exploration or dispersion under the control of the wandering threshold. The affinity threshold is employed under the achievement domain. Upon detecting a task during exploration, an AIS agent will evaluate its feasibility to tackle the task using the affinity threshold. The agent will either be activated or suppressed to approach its targeted task based on the affinity between the agent and the task. The activation threshold is employed under the cooperation domain. This threshold regulates the number of agent to participate in a cooperative task. Through receiving stimulation signals, a responding agent will either be activated or suppressed to join a cooperative task by the activation threshold. (Prasad & Rangaswami 2001)

In addition to the internal control of an AIS agent regarding individual level, control that ...