Spatial-temporal adaptive network partitioning for urban traffic signal control

Abstract In response to rapidly growing and diversifying traffic demand, it is necessary develop a network partitioning method that could achieve real-time global optimal performance adapt evolution. this paper, an adaptive presented, which achieves partitions at runtime determines appropriate Time-of-Day breakpoints update partition results simultaneously. For each time interval, schemes are firstly assessed in terms of modularity by taking roadway geometry, flow information, signal timing into account. Two values attained from the assessment: maximum obtained existing partition. Then updated, provided relative deviation these two exceeds given threshold for certain number successive intervals. Experimental show above-mentioned scheme outperforms some notable control techniques spatial aspect. temporal aspect, updating can well respond varying conditions yield significantly higher average modularity.