A tensor completion-based traffic state estimation model

Due to the low penetration rate and random routes of floating cars, the traffic state derived by the floating car system could not cover all the links of the road network, which has a great impact on the performance of dynamic traffic management and information applications. In this paper, the state of link which has the float car passing through is obtained by the float car system. For the link without passing through by the float car, a tensor form that can encode spatial and temporal correlation simultaneously is introduced to modeling the traffic state for the first time. Then, the traffic state of links without passing through by floating cars is completed by the tensor completion method which can well mines the multi-mode correlations of traffic state tensor model. In the experiment, the simulated data are used to evaluate the completion strategy in the tensor framework and the results show that the proposed strategy tends to deliver high-quality solutions.