Evaluating the performance of map matching algorithms for navigation systems: an empirical study

Singh, Jaiteg; Singh, Saravjeet; Singh, Sukhjit; Singh, Hardeep

doi:10.1007/s41324-018-0214-y

Cited by 29 publications

(6 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The point-to-curve method projects each point to the geometrically closest edge, while the curve-to-curve matching algorithms align each vehicle traveling trajectories with the closest network road links. Although these approaches are simple and intuitive, they are sensitive to noise, and connectivity problems exist during link transitions, resulting in unexpected results (Quddus et al, 2006;Singh et al, 2019;White et al, 2000). In Bouju et al (2002), authors reviewed the point-to-point and point-to-curve algorithms for map-matching processes in rural and urban areas, and pointed out three essential factors: historical data, topological information, and road design characteristics to improve the matching accuracy.…”

Section: Geometric Algorithmsmentioning

confidence: 99%

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Jin¹,

Choi²,

Yeo³

2021

Preprint

View full text Add to dashboard Cite

In many trajectory-based applications, it is necessary to map raw GPS trajectories onto road networks in digital maps, which is commonly referred to as a map-matching process. While most previous map-matching methods have focused on using rule-based algorithms to deal with the map-matching problems, in this paper, we consider the map-matching task from the data perspective, proposing a deep learning-based map-matching model. We build a Transformer-based map-matching model with a transfer learning approach. We generate synthetic trajectory data to pre-train the Transformer model and then fine-tune the model with a limited number of groundtruth data to minimize the model development cost and reduce the real-to-virtual gap. Three metrics (Average Hamming Distance, F-score, and BLEU) at two levels (point and segment level) are used to evaluate the model performance. The results indicate that the proposed model outperforms existing models. Furthermore, we use the attention weights of the Transformer to plot the map-matching process and find how the model matches the road segments correctly.

show abstract

Section: Geometric Algorithmsmentioning

confidence: 99%

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Jin¹,

Choi²,

Yeo³

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Finally, four different paths in Beijing were taken as examples, and the effectiveness and superiority of this method were verified. Singh et al [6] collected 47 km of track route data set, compared the map matching algorithms based on geometry, topology and Kalman filter, and found that the performance of the map matching algorithm based on Kalman filter was significantly better than the other two algorithms. In this study, the map matching algorithm for traffic route planning was briefly introduced, the part of the optimal path selection in the traditional map matching algorithm was improved by using Bayes theorem, and finally the two algorithms were simulated.…”

Section: Introductionmentioning

confidence: 99%

Study on the Rapid Planning of Traffic Routes Based on Map Matching Algorithm

2020

IJOMAM

View full text Add to dashboard Cite

With the development of positioning system, the data of vehicle trajectory that can be obtained by the traffic department is more and more huge, and it is more difficult to use the data of vehicle trajectory points and map to plan the traffic route. In order to improve the efficiency of traffic route planning, this paper briefly introduced the map matching algorithm for traffic route planning and improves the part of the optimal path selection in the traditional map matching algorithm by using Bayes theorem. The simulation results of the two algorithms showed that the trend of the two algorithms was basically the same, but the path of the improved algorithm was smoother, and the traditional algorithm had different degrees of distortion in the dense road network; the time interval of the track points provided by the positioning system affected the accuracy of the traffic route planning, the greater the interval, the lower the accuracy of the two algorithms, but under the same time interval, the accuracy of the improved algorithm was always higher than that of the traditional algorithm; the improved algorithm needed less operation time and had higher efficiency in the planning of traffic route than the traditional algorithm.

show abstract

“…As the essential data preprocessing step, the map-matching problem has been studied for more than two decades [14]. However, despite the massive number of map-matching algorithms proposed, only a few surveys [28,56,74,103,114,136] classify or compare them. Quddus [103] et al first summarised the early algorithms in 2007.…”

Section: Map-matching Algorithmsmentioning

confidence: 99%

“…Such categorisation was widely accepted but is now obsolete as it fails to classify most of the new methods proposed afterwards. Nevertheless, apart from some recent topic-specific surveys [56,74], this categorisation is still widely adopted by recent papers [73,101,118,122] and surveys [114], which indicates the need of a review in this field. Besides, recent works bring various new matching frameworks [112,119] and tuning techniques [57,63,81,95] to solve the map-matching problem on new types of positioning data (DGPS, inertial sensor, laser scanner [86], camera [71]) and new queries (lane-level map-matching [41,71,86]).…”

Section: Map-matching Algorithmsmentioning

confidence: 99%

See 1 more Smart Citation

A Study on map-matching and map inference problems

Chao¹

View full text Add to dashboard Cite

The recent upsurge of GPS(Global Positioning System)-equipped devices has enabled the tracking of users'/vehicles' locations. Meanwhile, the emergence of various location-based services emphasises the crucial role of the underlying digital map. However, the intrinsic inaccuracy of both maps and GPS trajectories has been a major issue that hinders the development of location-based applications, like navigation, location-based recommendation and traffic analysis. Several techniques are proposed to deal with the data quality issues in maps and GPS trajectories, respectively: (1) To avoid the excessive cost of ground surveying for map construction, the map inference algorithm aims to construct a digital map from GPS trajectories automatically. (2) To ensure the recency of digital maps, the map update algorithm focusses on updating an existing map using recent GPS trajectories. (3) To get rid family), Dr.

show abstract

Evaluating the performance of map matching algorithms for navigation systems: an empirical study

Cited by 29 publications

References 27 publications

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Transformer-based Map Matching Model with Limited Ground-Truth Data using Transfer-Learning Approach

Study on the Rapid Planning of Traffic Routes Based on Map Matching Algorithm

A Study on map-matching and map inference problems

Contact Info

Product

Resources

About