Urban Dynamic Traffic Assignment Model Based on Improved Ant Colony Algorithm

Abstract

The dynamic traffic assignment problem is a problem of finding the time-varying traffic volume on each directional section of the traffic network on the premise of knowing the topological structure of the urban traffic network and the time-varying traffic demand in the network. This problem is not only the premise of urban traffic control and guidance, but also the basis of urban traffic network toll system, and the key to urban traffic system planning and evaluation. This paper presents an improved ACA-based urban dynamic traffic assignment model. The model uses pseudo-random state transition rules and pheromone update rules of routes and road sections to simulate travelers' route selection behavior at road network nodes, and realizes the synthesis of static prior knowledge, dynamic traffic state and randomness of route selection in the process of route selection. The simulation results show that compared with the traditional ACA, the urban dynamic traffic assignment model based on the improved ACA can obtain better road network traffic balance, and it also has certain application value for the route guidance system under time-varying road conditions.