论文部分内容阅读
Floating car data are being increasingly applied in urban dynamic traffic data collection.However,the mobility and scale limit of floating cars may lead to incomplete or inaccurate data and thus influence the measurement of the state of traffic.To address the above problems,we use threshold control and a rule based on approximate normalization transform to clean abnormal traffic data,and fill the missing data by a weighted average method and an exponential smoothing method.The findings of data repair for road in Beijing demonstrate that the mean repair error may meet the requirements of traffic state measurement,indicating that this methodology effectively cleans floating car data.