Call number:
S 99.0139(342)
In:
Wissenschaftliche Arbeiten der Fachrichtung Vermessungswesen der Universität Hannover, Nr. 342
Type of Medium:
Series available for loan
Pages:
137 Seiten
,
Illustrationen, Diagramme
ISSN:
0174-1454
Series Statement:
Wissenschaftliche Arbeiten der Fachrichtung Vermessungswesen der Universität Hannover Nr. 342
URL:
http://www.gbv.de/dms/tib-ub-hannover/1030810273.pdf
Language:
English
Note:
Dissertation, Gottfried Wilhelm Leibniz Universität Hannover, 2018
,
1 Introduction
1.1 Background and motivation
1.2 Goals of this thesis
1.3 Outline
2 Background on digital maps and data mining
2.1 Digital maps
2.1.1 Navigation maps and map dynamics
2.1.2 OpenStreetMap
2.1.3 Navigation Data Standard (NDS)
2.2 Data mining
2.2.1 Knowledge Discovery in Databases (KDD) process
2.2.2 Taxonomy of data mining methods
2.2.3 Classification
2.2.4 Clustering
2.2.5 Time series analysis
3 Related work about mobile crowdsensing of on-street parking spaces
3.1 On-street parking
3.1.1 Parking occupancy detection
3.1.2 Parking availability estimation and prediction
3.1.3 Parking search and guidance
3.2 Mobile crowdsensing
3.2.1 Mobile crowdsensing in transportation
3.2.2 Mobile crowdsensing for parking
3.3 Research gaps addressed in this thesis
4 LiDAR-based parking availability data acquisition
4.1 Data recording
4.1.1 Sensor equipment
4.1.2 Measurement campaign
4.2 Methodology
4.2.1 Preprocessing
4.2.2 Segmentation
4.2.3 Classification
4.2.4 Repetition of segmentation and classification
4.2.5 Matching to road network
4.3 Results
4.3.1 Object segmentation
4.3.2 Classification
4.3.3 End-to-end evaluation of complete approach
4.3.4 Parking occupancy statistics over the day
4.4 Concluding remarks
5 Learning parking legality maps from parking observations
5.1 Methodology
5.1.1 Location of parked vehicles as method input
5.1.2 Data preprocessing
5.1.3 Definition of feature sets
5.1.4 Learning the parking legality of road subsegments
5.2 Evaluation
5.2.1 Evaluation approach
5.2.2 Results
5.3 Concluding remarks
6 Spatio-temporal analysis of large scale parking availability data and simulation of crowdsensing
6.1 Description and processing of parking dataset from SFpark
6.2 Time series analysis of parking availability data
6.3 Clustering of parking occupancy daily pattern
6.4 Spatial relations in parking availability
6.5 Modelling of crowdsensing based on downsampling for probe vehicles and mobile apps
6.5.1 Scenario based on probe vehicles
6.5.2 Scenario based on mobile apps
6.6 Modelling of probe-vehicle-based crowdsensing from taxi GPS trajectories
6.6.1 Processing overview and description of taxi trajectory dataset
6.6.2 Taxi GPS trajectory processing
6.6.3 Characteristics and aggregation of taxi coverage
6.6.4 Comparison of parking and taxi daily pattern
6.6.5 Simulation of parking availability observations
6.7 Concluding remarks
7 Parking availability estimation and prediction from crowdsensed data
7.1 Spatial interpolation of parking availability
7.2 Parking availability estimation with persistence method
7.3 Estimation and prediction of parking availability based on binary classification
7.3.1 Binary classification approach
7.3.2 Results of binary classification estimation and prediction
7.4 Concluding remarks
8 Benefits of crowdsensed parking availability information
8.1 Types of information for on-street parking
8.2 Experimental setup
8.2.1 Routing strategies
8.2.2 Data sources
8.3 Evaluation of the impact of different parking information
8.3.1 Results for all decisions in the dataset
8.3.2 Results for relevant decisions
8.3.3 Similarity of capacity
8.4 Concluding remarks
9 Conclusion and outlook
9.1 Research questions addressed and overall conclusion
9.2 Applicability of dynamic map approaches to further dynamic phenomena
9.3 Future research directions
List of figures
List of tables
References
Acknowledgements
Curriculum vitae
Location:
Lower compact magazine
Branch Library:
GFZ Library
