Make Driver Agent More Reserved: An AIM-Based Incremental Data Synchronization Policy

AIM is one of the leading Autonomous Intersection Management mechanisms based on Multiagent System (MAS) for alleviating traffic congestion specially at intersections. One of the concerned problems on AIM, however, lies in the communication complexity of the system. Previously, the driver agent has no choice, but to completely retransmit its adjusted request information when the former reservation is rejected by the intersection manager, which results in the increase of interaction complexity between agents and the plenty of redundant data transmission. In this paper, we present an incremental data synchronization policy ksync for driver agent to avoid such redundant retransmission. In particular, we first introduce the basic properties of ksync policy. Second, we demonstrate how ksync could be well integrated into the knowledge base of driver agent as one of its essential policies. Third, we prove by experimental evaluation that the average data compression rate can be improved by over 80% exploiting ksync. Finally, we propose some of the most significant research prospects on ksync using the techniques in data mining and machine learning.