Multi-agent based vehicular congestion management

Abstract

In rapidly growing transportation networks, traffic congestion can result from inefficient traffic control infrastructure or ineffective traffic control measures. Existing congestion management techniques in Intelligent Transportation Systems (ITS) have not been very effective due to lack of autonomous and collaborative behavior of the constituent traffic control entities involved in these techniques. Moreover, these entities cannot easily adapt to the traffic dynamics and the traffic control intelligence is mostly centralised making it susceptible to overload and failures. The autonomous and distributed nature of multi-agent systems is well-suited to the transportation domain which is dynamic and geographically distributed. This paper reviews existing congestion management techniques and discusses their limitations. The paper, further, comprehensively surveys multi-agent techniques for congestion management in ITS and describes their advantages over other existing techniques. The paper classifies the multi-agent techniques based on the locus of decision control intelligence and focuses on their suitability of application in congestion management. We conclude with outstanding issues and challenges.