信号交叉口行人群集分类方法研究

以信号交叉口行人群体为研究对象,借鉴生物界群集研究的成果,给出了行人群集的定义,同时提出群集半径的概念和基于搜索的群集半径计算方法。通过对北京市信号交叉口60个小时视频资料的737个行人群集样本的分析,得到不同行人群集的特征参数值。研究结果表明,行人群集半径能够较好地评价我国城市信号交叉口人行横道的服务水平,可以为信号交叉口大量行人过街的控制算法和模型的研究奠定基础。

Classification Method Research of Pedestrian Swarming at Signalized Intersection

Taking the pedestrian groups at a signalized intersection as a research object, and using the achievements of living things for reference, the conception of pedestrian swarming was given. At the same time, the conception of pedestrian swarming radius was defined, and a method based on searching was provided too. By analyzing 737 pedestrian swarming samples of 60 hour video data at a signalized intersection, the feature parameters of different pedestrian swarmings were obtained. The research result showed that pedestrian swarming circus could evaluate the service level of the pedestrian crossing at the signalized intersections of China, and the researches of the control algorithm and model of pedestrian crossing were founded.