Abstract:

To address the problem that in the complex urban environment, due to the inevitable interruption of GNSS positioning signal and the accumulation of errors during vehicle driving, the collected vehicle trajectory data was likely to be inaccurate and incomplete.a bidirectional weighted trajectory reconstruction algorithm was proposed based on RNN neural network.The GNSS-OBD trajectory acquisition device was used to collect vehicle trajectory information, and multi-source data fusion was adopted to achieve bidirectional weighted trajectory reconstruction.Furthermore, the neural arithmetic logic unit (NALU) was leveraged with the purpose of enhancing the extrapolation ability of deep network and ensuring the accuracy of trajectory reconstruction.For the evaluation, real-world experiments were conducted to evaluate the performance of the proposed method in comparison with existing methods.The root mean square error (RMSE) indicator shows the algorithm accuracy and the reconstructed trajectory is visually displayed through Google Earth.Experimental results validate the effectiveness and reliability of the proposed algorithm.

Key words: private car, vehicle positioning, trajectory reconstruction, RNN