Research On Spatio-Temporal Trajectory Query And Route Planning Based On Activity Trajectory Methods

With the widespread use of GPS-enabled devices such as vehicles and smartphones,massive amounts of spatio-temporal trajectory data are being generated from location-based services.These data contain rich information.Therefore,efficient storage,com-pression and query of spatio-temporal trajectory data will have a profound impact on ex-isting traffic,economy and environment.Meanwhile,there are many mature applications of path length-based and trajectory-based path planning methods,and it is still a challenge to use activity trajectories to support path planning.For spatio-temporal trajectory query and path planning,researchers have proposed many methods in recent years,however,the current algorithms still have shortcomings in specific aspects.In this paper,we propose a solution to this problem,which consists of four main parts as follows.First,a multi-granularity spatio-temporal trajectory organization method based on k~2tree is proposed.In this paper,we propose a multi-granularity data model for spatio-temporal trajectories,which depicts the spatial structure and temporal changes of spatio-temporal trajectories at different levels.Based on this,a k~2tree based spatio-temporal trajectory data organization method is designed to realize the integration and association of spatio-temporal trajectories at the data organization level,which improves the com-pression and retrieval performance of spatio-temporal trajectories.Second,a spatio-temporal trajectory indexing and query processing algorithms is proposed.In order to improve the trajectory data management capability and build a bridge between spatio-temporal trajectory data and upper-level analysis applications,we designs a spatio-temporal multi-version trajectory index and supports multiple query pro-cessing algorithms.Firstly,the multi-granularity spatio-temporal trajectory organization method based on k~2tree is adopted to process spatio-temporal trajectory data can well save data storage space and support querying under compressed data.Next,the corre-sponding query optimization algorithms are designed for different query types.Finally,experiments on the real trajectory datasets show that the proposed method has the char-acteristics of efficient query and good scalability.Third,a query algorithm of spatio-temporal trajectory similarity based on graph structure is proposed.For the problem of slow spatio-temporal trajectory similarity query under spatial constraints,we propose a graph structure-based spatio-temporal trajectory similarity query algorithm.This paper designs a trajectory similarity metric function that synchronously matches spatial and temporal distances,and uses a backward index struc-ture that incorporates temporal and spatial filtering to reduce computational costs and effectively improve query efficiency.Finally,experiments on real datasets demonstrate the effectiveness and accuracy of the proposed algorithm.Fourth,a path planning algorithm based on user activity trajectories is proposed.In response to the lack of consideration of user activity intent in existing work,we propose a path planning method that takes into account user activity intent and path length.First,the spatial coordinates and semantic information of POIs are extracted from user activity tra-jectories to construct an information network fusing POI information and road networks.Second,the improved aggregated R-tree is used to introduce user demand diversity into the greedy algorithm to search for paths in road networks that meet user activity intentions and have shorter paths.Finally,experiments on real data sets demonstrate the effective-ness of the proposed algorithm.