GNSS for Train Localization Trajectory Generation Featuring Depth-First-Search Method

Global Navigation Satellite Systems (GNSS) for train localization does not rely on trackside equipment, which realizes “trainborne centric” positioning using onboard localization sensors. It is one of the important methods for the future train operation & control system as an advancement for train localization. European Next Generation of Train Control (NGTC) and Chinese Dynamic Spacing Train operation & control system (DTCS) all apply GNSS and Digital Track Map (DTM) together for train localization. Before field test of the train localization, it is necessary to establish a GNSS test environment, generate train trajectories, carry out important research contents of laboratory simulation testing of train operation & control system functions and performance in-lab-test. Based on the DTM, this paper generates a topology model includes point of interests (POIs) and track pieces relationships. Considering train operation safety constrains and track constrains, using the depth-first-search (DFS) and dual-stack data structure, all possible paths from origin and destination (OD) are generated as a path candidate set, the consistency of safe route in the path candidate set is performed, and the safe and available path is validated. Then the path geographical information is generated using DTM. The Haergai-Muli Railway in Qinghai is used in this paper to investigate the method, 6 planned trajectories are generated. The generated trajectory is verified by GNSS simulator, and the 95th quantile error between the received data and the original data is 0.62 m, which meets the DTCS accuracy requirement.