Parking Availability Detection via Vehicle Trajectory Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems for providing parking availability information are inefficient in city centers and residential areas, as they mainly rely on data from journey start and end points, leading to data sparsity and less accurate or timely information, especially with limited vehicle participation.
Innovation Solution
A system that processes vehicle trajectory data to identify initiation points of parking searches and classifies street segments as full or unavailable, providing parking availability information for segments beyond start and end points, using trajectory data from vehicles equipped with sensors and a trajectory processing platform that aggregates and analyzes this data to offer more precise guidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If parking availability information is collected only from journey start and end points, then data collection is simple, but information accuracy and coverage are insufficient
Solution Approach 1:
The patent segments the journey into multiple portions and identifies parking search initiation points within each segment. By analyzing trajectory data at multiple points along the journey rather than only at start and end points, the system divides the data collection process into manageable segments, improving parking availability information accuracy while maintaining systematic simplicity
Solution Approach 2:
The patent adds a temporal and spatial dimension to data collection by identifying when and where drivers initiate parking searches during their journeys. This transforms the traditional two-point (start/end) data collection into multi-point trajectory analysis, enhancing information coverage and accuracy without proportionally increasing system complexity
2Quantity of substance
If more vehicle trajectory data is processed to improve parking information coverage, then data sparsity is reduced, but processing resources and time increase
Solution Approach 1:
The patent extracts only the critical information needed for parking availability detection from vehicle trajectory data - specifically identifying when and where parking searches are initiated. By extracting only these essential data points rather than processing entire trajectory datasets, the system increases effective data points for parking analysis while minimizing processing time and resource consumption
Solution Approach 2:
The patent applies partial action by focusing analysis on specific portions of journeys where parking searches are most likely to occur. Rather than uniformly processing all trajectory data, the system selectively analyzes segments containing parking search initiation points, efficiently increasing data coverage without proportional increases in processing time
3Loss of information
If parking search initiation points are identified throughout the journey, then parking availability information coverage is improved, but analysis complexity increases
Solution Approach 1:
The patent performs preliminary classification of trajectory data to identify portions associated with parking searches before detailed analysis. By pre-identifying parking search initiation points through deviation detection from reference routes, the system reduces the complexity of subsequent analysis while ensuring comprehensive parking availability information coverage
Solution Approach 2:
The patent introduces reference routes as an intermediary element to simplify the identification of parking search initiation points. By comparing actual vehicle trajectories against predetermined reference routes, the system creates a clear criterion for detecting parking search behavior, reducing analysis complexity while improving information completeness
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
An approach is provided for parking availability detection based on vehicle trajectory information. A trajectory processing platform processes and/or facilitates a processing of trajectory data associated with at least one journey of at least one vehicle to determine at least one portion of the at least one journey that is associated at least one parking search by the at least one vehicle. The trajectory processing platform also determines one or more street segments associated with the at least one portion of the at least one journey. The trajectory processing platform further causes, at least in part, a classification of the one or more street segments as associated with the at least one parking search. The trajectory processing platform further determines parking availability information for the one or more street segments based, at least in part, on the trajectory data.