1 / 27

An Interactive-Voting Based Map Matching Algorithm

An Interactive-Voting Based Map Matching Algorithm. ——By Katrina Zeng. 1. Introduction. 2. Related Work. 3. Preliminary. 4. Algorithm & Algorithm Analysis. 5. Conclusion. Contents. Introduction. What is map matching?

thalia
Download Presentation

An Interactive-Voting Based Map Matching Algorithm

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. An Interactive-Voting Based Map Matching Algorithm ——By Katrina Zeng

  2. 1. Introduction 2. Related Work 3. Preliminary 4. Algorithm & Algorithm Analysis 5. Conclusion Contents

  3. Introduction • What is map matching? Match an original GPS tracking data to a digital map or a digital road network. • What is map matching algorithm? Identify the true road segment on which a user (or a vehicle) is/was travelling. • What is the challenge of map matching algorithm? Low-sampling-rate GPS tracking data.

  4. Introduction Distribution of the sampling interval

  5. Introduction • What is the problem this paper solved? Itproposed an adaptive and robust map matching algorithm with acceptable accurate for map matching low- sampling-rate GPS tracking data. • What are the key insights of this paper? 1) Position Context 2) Mutual Influence 3)Weighted Influence • What are the contributions of this paper? 1) IVMM algorithm 2) Extensive experiments on real datasets 3) The evaluation of the algorithm

  6. Introduction Illustration of the insights

  7. 1. Introduction 2. Related Work 3. Preliminary 4. Algorithm & Algorithm Analysis 5. Conclusion Contents

  8. Related Work • Involved Information of Input Data Geometric: Geometric map matching algorithms utilize the geometric information of the spatial road network data by consider only the shape of the links without the connectivity of the links. Topological: Topological methods use the topology of map features to constrain the candidate matches for a sampling point. The approach in this paper is based on both geometric and topological information of the road network.

  9. Related Work • Global Algorithms and Local Algorithms Local/Incremental algorithms:follow a greedy strategy of sequentially extending the solution from an already matched portion. Global algorithms: find a trajectory which is as closer as the sampling track among all available trajectories in the road network This paper adopted global algorithm considering the mutual influence and the impact of remoteness on the positions of the sampling points.

  10. Related Work • Sampling Rate Dense-sampling-rate approaches Low-sampling-rate approaches This paper adopted a low-sampling-rate algorithm.

  11. 1. Introduction 2. Related Work 3. Preliminary 4. Algorithm & Algorithm Analysis 5. Conclusion Contents

  12. Preliminary • Problem Definition Definition 1. GPS trajectory: p1 → p2 → p3 …. → pn (pi.lat, pi.lng, pi.t) Definition 2. Road network: graph G(V,E) V:terminal points of the road segments E:road segment e (e.eid, e.v, e.l, e.start, e.end) Definition 3. Path: e1 → e2 → e3 … → en (e1.start = Vi , en.end = Vj , ek.end = ek+1.start, 1 ≤ k ≤ n ) The problem of this paper is that find a path in G which matches T with its real path.

  13. Preliminary • IVMM Algorithm IVMM algorithm which is aimed to make the best of the interactive relationship among all the sampling points so as to find a global optimal path to match the trajectory, especially for low-sampling-rate situation

  14. 1. Introduction 2. Related Work 3. Preliminary 4. Algorithm 5. Conclusion Contents

  15. Algorithm • Candidates Preparation

  16. Algorithm • Candidates Preparation

  17. Algorithm • Position Context Analysis measurement error: construct a candidate graph G’(V’,E’):

  18. Algorithm • Position Context Analysis spatial analysis function: temporal analysis function:

  19. Algorithm • Mutual Influence Modeling Static Score Matrix Building:

  20. Algorithm • Weighted Influence Modeling Static Score Matrix Building:

  21. Algorithm • Weighted Influence Modeling Distance Weight MatrixWi:

  22. Algorithm • Weighted Influence Modeling Weighted Score Matrix:

  23. Algorithm • Interactive Voting Algorithm 1 Interactive Voting:

  24. Algorithm • Interactive Voting Algorithm 2 FindSequence:

  25. 1. Introduction 2. Related Work 3. Preliminary 4. Algorithm 5. Conclusion Contents

  26. Conclusion • IVMM algorithm • Extensive experiments on real datasets • The evaluation of the algorithm

  27. Thank you!

More Related