Signal control for multi-intersections is a hot issue in traffic research. Intelligent Vehicle and Infrastructure Coordination System (IVICS) helps us to develop much more accurate and effective signal control strategy. This study presents a system with a new algorithm to optimize the control strategies. This system collects the data in real time and develops strategies in the form of game tree. The evaluation algorithm for the control strategy of each intersection is put forward. And then a new regional coordinated signal control strategy is proposed based on game tree theory. The global optimal strategy can be obtained from the gaming of all the intersections. In other words, all the intersections acquire the optimal set of control strategies when each intersection acquires their own optimal strategy. This study utilizes VISSIM to verify the proposed algorithm. Compared with that of the fixed cycle control system, the simulation study implemented in a road network with five intersections shows that the proposed algorithm is effective in coordinating multiple intersections, especially when the traffic flow density is mean or close to the saturation.
展开▼
