A novel automatic track identification algorithm based on LTS-Hausdorff distance
Xi-Hui Yan, Jian Wang, Baigen Cai, Wei Shangguan · 2011
Train locating information is very important for train control system, how to achieve automatic identification of train track occupancy using GNSS simply in some railway sections without track circuits has been a crucial problem. Hausdorff distance can be used to measure the mismatch between two sets, which is widely used in object matching. This paper presents a novel algorithm for automatic identification of train track occupancy based on LTS-Hausdorff distance and D-S evidence theory. The LTS-Hausdorff distance reference template of railway track was established, the calculation process of LTS-Hausdorff distance and the identify strategy of train track occupancy based on D-S evidence theory were studied. Test results show that the algorithm is efficient and can achieve automatic track identification in low cost.