LAUSR

The traditional transportation system is based on a fixed-timed strategy to control the traffic congestion on urban roads. However, the increase of vehicle density in a smart city implies the variation and the conflict on the demand pattern of the drivers. The smart city development requires to create an efficient control plan for an intelligent traffic management system. In this paper, we aim to reduce the congestion at signalized intersections, and satisfy the needs of drivers according to their degree of displacement urgency. We use two optimization methods, namely the synchronization and the genetic algorithm ($$\hbox {GA}$$GA), where we develop three scheduling protocols. The first is the intelligent context-aware negotiation protocol ($$\hbox {ICANP}$$ICANP). This protocol allows the negotiating vehicles to cross the intersection. It enables each traffic light at the signalized intersection to negotiate the green time assigned to its phase. $$\hbox {ICANP}$$ICANP uses $$\hbox {GA}$$GA to optimize the crossing time in order to minimize the total waiting time of negotiating vehicles. Moreover, we introduce a negotiation protocol based on reputation, which minimises the congestion effect from incoming dishonest drivers. Finally, we propose an intelligent context-aware priority protocol ($$\hbox {ICAPP}$$ICAPP), that considers the existence of priority vehicles. Upon arrival of at least one priority vehicle at the signalized intersection, $$\hbox {ICAPP}$$ICAPP interrupts the green time of negotiating vehicles. A series of simulation showed that the proposed protocols reduce the total waiting time and the emissions of $$\hbox {CO}_{2}$$CO2 of vehicles at signalized intersection, in comparison with circular, $$\hbox {ITLC}$$ITLC and $$\hbox {CATLS}$$CATLS scheduling algorithms. Furthermore, formal complexity analysis and performance evaluation show the effectivity of our protocols.