摘要
针对交通流的特点,建立了基于神经网络的交通状态模糊判别方法。综合考虑检测器采集的流量、速度和占有率信息,采用三个模糊规则进行推理,利用具有模糊输出的BP神经网络对交通状态进行评价。利用虎门连升路采集的交通信息对算法进行了验证。研究表明,该方法具有较强的自学习、自组织和自适应能力,不仅可以确定交通的状态,而且可以识别出属于该状态的程度,使判别结果更加具体,为交通状态的判别提供了一种新思路。
According to the features of traffic flow,a traffic condition fuzzy recognition method based on neutral network was proposed.BP neural network was used to realize the traffic condition fuzzy recognition by considering the traffic speed,flow and occupancy.The result shows that the proposed method has excellent self-learning,self-organizing and adaptive abilities,which not only determines the traffic state,but also identifies the degree of belonging to the state.Compared with the common method,the recognition result is more specific,providing a new idea for traffic recognition.
出处
《科学技术与工程》
2010年第21期5195-5199,共5页
Science Technology and Engineering
基金
国家863计划项目(2006AA11z211)资助
关键词
交通状态
神经网络
模糊判别
traffic state neural network fuzzy recognition
