基于毫米波雷达的城市轨道交通乘客统计分析方法研究

Research on the statistical analysis method of urban transit passengers based on millimeter wave radar

乘客上下车人数实时统计是实现智慧城市轨道交通的关键环节,然而目前大多基于光学的视频统计方法在复杂的高密度人流场景下存在一定缺陷。文章提出一种创新的基于毫米波雷达的乘客上下车计数方案,利用快速傅里叶变换和恒虚警率提取乘客的折射点距离、方位角和速度信息。为解决高密集人流条件中的遮挡问题,引入基于密度的空间聚类算法,可在拥挤环境中有效识别并区分不同的乘客群体。为实现乘客目标的连续动态跟踪,提出一种基于卡尔曼滤波的智能跟踪技术,该技术可确保即使在人流快速变动的情况下,也能稳定地跟踪目标乘客。通过一系列算例验证,证明该方案的有效性。车站模拟测验证明,方案在密集人流条件下仍展现良好的跟踪准确性。

Real-time statistics on the number of passengers getting on and off the train is a key link in the realization of smart rail transit.However,most contemporary optical-based video statistics methods have certain defects in complex and high-density crowd flow scenes.In this regard,this article proposed a solution for counting passengers on and off trains based on millimeter wave radar.Firstly,the distance,azimuth and velocity of the refractive point of passengers were extracted by FFT(Fast Fourier Transformation)and CFAR(Constant False-Alarm Rate).Secondly,DBSCAN(Density-Based Spatial Clustering of Applications with Noise)was introduced to solve the occlusion issue under the condition of dense people flow,which is able to effectively recognize and differentiate different groups of passengers in crowded environments.Thirdly,to realize the continuous dynamic tracking of passenger targets,an intelligent tracking technique based on Kalman filtering is proposed,which could ensure the stable tracking of the target passengers even in the case of rapid changes in people flow.The effectiveness of the scheme is demonstrated through a series of arithmetic validations.Finally,in the station simulation test,the scheme still demonstrates high tracking accuracy under dense crowd conditions.