Traffic State Estimation via a Particle Filter with Compressive Sensing and Historical Traffic Data

摘要

In this paper we look at the problem of estimatingudtraffic states within segments of road using a particle filter andudtraffic measurements at the segment boundaries. When there areudmissing measurements the estimation accuracy can decrease. Weudpropose two methods of solving this problem by estimating theudmissing measurements by assuming the current measurementsudwill approach the mean of the historical measurements from audsuitable time period. The proposed solutions come in the formudof an l1 norm minimisation and a relevance vector machine typeudoptimisation. Test scenarios involving simulated and real dataudverify that an accurate estimate of the traffic measurements canudbe achieved. These estimated missing measurements can then beudused to help to improve traffic state estimation accuracy of theudparticle filter without a significant increase in computation time.udFor the real data used this can be up to a 23.44% improvementudin RMSE values.