Traffic Speed Estimation Using Spatial Temporal GPS Smoothing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for traffic speed estimation across transportation networks face challenges due to the sparseness of GPS data, which is incomplete and does not record direction of travel, making it difficult to accurately estimate speed across inter-connected network segments.
Innovation Solution
A system and method that applies spatial and temporal smoothing processes to GPS data to identify and fill in missing speed values, using data processing techniques to model GPS data from multiple sources and provide real-time traffic speed estimates for all segments at all times by snapping GPS data to road links, building rescaled speed profiles, and applying clustering analysis for efficient storage and inference.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If GPS data is collected continuously at multiple locations to estimate traffic speed across large geographic areas, then coverage area is improved, but data completeness deteriorates because GPS data is only available for part of the links and only for part of the time
Solution Approach 1:
The system performs preliminary actions by collecting GPS data continuously at multiple locations before the actual estimation is needed. This advance data collection ensures that when speed estimation is required, the necessary data points are already available, allowing for more accurate and complete speed estimates across the entire transportation network without requiring real-time data collection at every moment.
Solution Approach 2:
The patent introduces intermediary processing steps that include direction-of-travel determination and link matching algorithms. These intermediaries bridge the gap between raw GPS data and complete link speed estimates by inferring missing information through spatial relationships and network topology, thereby filling in the data completeness gap while maintaining broad coverage.
2Area of stationary object
If GPS data is used to estimate speed across inter-connected network segments, then network coverage is improved, but measurement precision deteriorates because GPS data does not record direction of travel and there is interference between segments on unrelated routes
Solution Approach 1:
The system segments the transportation network into distinct links and road segments, then processes GPS data separately for each segment while maintaining context about direction of travel. This segmentation allows precise speed estimation for each link without interference from unrelated routes, while still providing comprehensive network coverage through the aggregated results.
Solution Approach 2:
The patent applies asymmetric processing by determining direction of travel for each GPS data point and using this directional information to match GPS observations to specific network links. This asymmetric approach to data processing enables precise speed estimation for each directed link while maintaining the ability to cover the entire network, resolving the contradiction between coverage and precision.
3Measurement precision
If complete link speed estimates are obtained by collecting GPS data at all locations and times, then measurement precision is improved, but data collection complexity and storage requirements increase
Solution Approach 1:
The system extracts only the necessary information from GPS data points - specifically speed, location, and direction of travel - while discarding redundant information. This extraction process achieves complete link speed estimates with reduced data complexity, as only the essential parameters needed for accurate speed estimation are collected and processed, rather than maintaining all possible GPS data.
Solution Approach 2:
The patent transforms raw GPS data into processed speed estimates by changing the parameters of the data representation. Instead of storing complete GPS trajectories, the system converts this data into link-level speed estimates with associated confidence metrics. This parameter transformation achieves high measurement precision while significantly reducing storage requirements and system complexity.
Data Source
AI summary
Estimation of traffic speed includes applying data processing functions to determine missing speed information by smoothing spatial and temporal GPS data to achieve an accurate estimation of link speed over all links of a transportation network at all time periods. This estimation of traffic speed uses one link's observed speed information to estimate neighboring links without observed speed information and therefore provides a system and method of processing collected GPS data to obtain a thorough understanding of traffic flow conditions for all represented links without further collection of GPS data. The present invention also provides a framework for analyzing and improving real-time collection of GPS speed data.