Destination prediction method and system based on hierarchical space-time embedded transformer model

By adopting a hierarchical spatiotemporal embedded transformer model in destination prediction, combining grid division, spatiotemporal encoding and feature pyramid transformer FPT, the problems of sparse data and insufficient multi-scale feature extraction in the existing methods are solved, and more efficient and accurate destination prediction is achieved.

CN119646321AActive Publication Date: 2025-03-18YIYANG XINDA TIANMA ELECTRICAL EQUIP MFG CO LTD
4 Cites 0 Cited by

Patent Information

Application Number
CN202411470181.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-21
Publication Date
2025-03-18
Estimated Expiration
2044-10-21

AI Technical Summary

Technical Problem

Existing neural network-based destination prediction methods face problems such as sparse data, missing time information and insufficient multi-scale feature extraction, resulting in poor prediction accuracy and efficiency.

Method used

Using a method based on a hierarchical spatiotemporal embedding transformer model, the spatiotemporal information and multi-scale features in the trajectory data are extracted through coarse-grained and fine-grained grid division, spatiotemporal coding and feature pyramid transformer FPT, and the destination prediction is performed in combination with a convolutional neural network and a Transformer encoder.

Benefits of technology

It effectively solves the problems of sparse data, missing time information and insufficient multi-scale feature extraction, improves the accuracy and efficiency of destination prediction, and provides more refined and personalized travel destination prediction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a destination prediction method and system based on a hierarchical space-time embedded transformer model, and the method comprises the steps: carrying out the coarse-grained mesh division and space-time coding of a region corresponding to a part of trajectory # imgabs0 #, obtaining a coarse-grained space-time trajectory image representing the general trend of the trajectory, and carrying out the prediction and extraction of a destination mesh region through a CNN (Convolutional Neural Network); the method comprises the following steps: extracting GPS track points, performing fine-grained grid division and space-time coding on a destination grid region to obtain a refined destination region track image, extracting motion features of the GPS track points in the destination region by using a feature pyramid transformer FPT, predicting the probability of candidate destinations by using a full-connection network and a Softmax classifier, and determining the output candidate destinations. The invention aims to extract the spatio-temporal information from the trajectory data to predict the destination, and solves the problems of data sparsity, time information loss and insufficient multi-scale feature extraction in the existing method.
Need to check novelty before this filing date? Find Prior Art