Proposal of adaptive coordination formation control mechanism for multiagent planning

本文研究了自适应多智能体规划队形变换算法。集中型、直接协调型和间接协调型是典型的形态。本文主要以基于多智能体的交通控制系统为研究对象，提出了基于多智能体的智能交通系统红绿灯控制框架。为了使交通流畅，交通灯的实时自适应协调是必要的，但许多传统的方法是集中控制型的，不具有这一特性。我们的基于多智能体的控制框架结合了间接协调和直接协调。通过间接协调实现对动态交通流的反应，通过直接协调实现绿波形成，即多个红绿灯参与的系统交通流控制策略。我们给出了该框架的详细机制，并通过仿真进行了对比评估，验证了其有效性。

In this study we have studied adaptive multiagent planning formation change algorithm. Centralized type, direct coordination type, and indirect coordination type is typical formation. This time, we have mainly adopted multiagent based traffic control system as case study, and proposed multi agent based traffic light control framework for intelligent transport systems. For smooth traffic flow, real-time adaptive coordination of traffic lights is necessary, but many conventional approaches are of the centralized control type and do not have this feature. Our multi agent based control framework combines both indirect and direct coordination. Reaction to dynamic traffic flow is attained by indirect coordination, and green-wave formation, which is a systematic traffic flow control strategy involving several traffic lights, is attained by direct coordination. We show the detailed mechanism of our framework and verify its effectiveness through comparative evaluation through simulation.