Trajectory Map Matching Methods Based on Stream Computing and Mutual Voting

In response to the problems of poor trajectory matching accuracy, high matching cost, limited map accuracy, and inability to handle large-scale trajectory data in current traditional map matching methods, this paper conducted research on trajectory map matching based on stream computing and mutual voting. Extracting trajectory data features through stream computing can better explain the characteristics and changes of trajectory data, improving the accuracy of large-scale trajectory data research. Mutual voting was used to study map matching, effectively processing continuously updated trajectory data, and utilizing information between similar trajectories for map matching. The map matching accuracy of the method studied in this article was over 95.9%. After 20 experiments, the average matching accuracy was 97.66%. Research on trajectory map matching methods based on streaming computing and mutual voting has improved the accuracy and real-time performance of map matching.