基于支持向量机的城市道路交通拥堵判别算法研究

近年来,城市交通拥堵现象越发严重,研究分析了国内外大量关于城市交通拥堵界定与判别的基础上,基于模式识别理论中支持向量机分类算法设计提出一种"畅行"、"一般拥堵"及"严重拥堵"道路拥堵三分类研究模式。以南京市虎踞路这一城市主干道路段为算法实例研究对象,结合实测采集和Vissim仿真拥堵交通流数据,借助Matlab实现设计算法的城市道路拥堵分类和判别,实验结果表现出较好的分类和检测效果,表明设计算法应用城市拥堵判别是可行的,且可以进一步优化提高。

The Detection Algorithm Research on Urban Road Traffic Congestion Based on Support Vector Machine

In recent years, the phenomenon of urban traffic congestion is becoming more seriuus. After analyzing a large number of domestic and foreign research on urban traffic congestion and diseriminant, support vector machine classification algorithm which is on the basis of the of pattern recognition presents a classification mode including ＂ Hang＂ , ＂general congestion＂ and ＂ serious congestion＂ road congestion. This passage took the eity of Nanjing Hu Ju Road as the main road link to algorithm instance object of studv, with the collection and Vissim Simulation congestion traffic flow data, using Matlab to achieve the desigu algorithm urban road congestion classification and identification. The experimental results showed a better lassil＇ieation and lest results. It indicates that the design algorithm is feasible to be applied in urban congestion determination, and it can be further optimized to improve.