Robust estimation of traffic density with missing data using an adaptive-R extended Kalman filter
作者：
Highlights：
• This paper proposes a novel adaptive-R extended Kalman filter (AREKF) combined with a model-based data imputation technique to estimate traffic density even when data is missing, due to for example sensor failure.
• We show analytically that the AREKF is able to accurately estimate the density even when the noise covariance matrices are not accurately known; as a result, it is no longer necessary to accurately know the noise covariance matrices which are often unknown and varying with time.
• The estimated density from the AREKF with data imputation is fed into a real-time ramp metering control algorithm to control vehicle flow on a merging road, which is highly susceptible to traffic congestion; the results show that the AREKF can accurately estimate the traffic density, and the ramp metering control algorithm yields a significant improvement to the traffic flow and thus, alleviates congestion.
• The computational cost of our proposed method with normal data, missing data, and missing data imputation is very low, which makes it suitable for real-time implementation.
摘要
•This paper proposes a novel adaptive-R extended Kalman filter (AREKF) combined with a model-based data imputation technique to estimate traffic density even when data is missing, due to for example sensor failure.•We show analytically that the AREKF is able to accurately estimate the density even when the noise covariance matrices are not accurately known; as a result, it is no longer necessary to accurately know the noise covariance matrices which are often unknown and varying with time.•The estimated density from the AREKF with data imputation is fed into a real-time ramp metering control algorithm to control vehicle flow on a merging road, which is highly susceptible to traffic congestion; the results show that the AREKF can accurately estimate the traffic density, and the ramp metering control algorithm yields a significant improvement to the traffic flow and thus, alleviates congestion.•The computational cost of our proposed method with normal data, missing data, and missing data imputation is very low, which makes it suitable for real-time implementation.
论文关键词：AREKF,Data imputation,Ramp metering,Traffic congestion,Traffic density estimation
论文评审过程：Received 7 August 2021, Revised 7 December 2021, Accepted 31 December 2021, Available online 14 January 2022, Version of Record 14 January 2022.
论文官网地址：https://doi.org/10.1016/j.amc.2022.126915
