Vehicle Trip Classification via Geographic Tile Fingerprinting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems lack an efficient method to classify vehicle trips based on telematics data, which hinders the processing and analysis of vehicle operation data.
Innovation Solution
A system and method that classify vehicle trips by obtaining geographic location points, identifying geographic tile identifiers, deriving a fingerprint value, and assigning a classification label by comparing the fingerprint value to a database of previous trips.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If telematics data is processed without trip classification, then processing speed is maintained, but data analysis accuracy and insight quality deteriorate
Solution Approach 1:
The system segments trips into different categories (commute trips, errand trips, leisure trips) based on geographic location analysis. By dividing the overall telematics data processing into trip-specific segments, the system can apply targeted analysis methods to each segment, improving overall data analysis accuracy while maintaining manageable processing complexity through organized categorization.
Solution Approach 2:
The patent introduces trip classification as an intermediary layer between raw telematics data collection and detailed data analysis. This intermediary classification step organizes data by trip type before analysis, enabling more accurate insights without requiring complete reprocessing of the entire data pipeline, thus balancing accuracy improvement with system complexity constraints.
2Loss of information
If detailed trip classification is implemented, then insight quality improves, but processing time increases
Solution Approach 1:
The system performs preliminary trip classification by analyzing geographic location data to determine trip types before conducting detailed telematics analysis. This preliminary action of categorizing trips upfront allows subsequent analysis to focus on trip-specific patterns, improving information quality while reducing overall processing time by avoiding generic analysis of all trips uniformly.
Solution Approach 2:
The patent applies different analysis methods and criteria to different trip types (commute, errand, leisure) based on their specific characteristics. By tailoring the analysis approach to each trip category rather than using a uniform method, the system extracts higher quality insights from each trip type while optimizing processing efficiency for each category's specific requirements.
3Measurement precision
If geographic location points are collected for all trips, then trip classification accuracy improves, but data storage requirements increase
Solution Approach 1:
The system extracts only the essential geographic location information needed for trip classification (such as start location, end location, and key途经 points) rather than storing all raw location data points. By taking out only the necessary geographic elements required for accurate trip categorization, the system achieves high classification accuracy while minimizing data storage requirements.
Solution Approach 2:
The patent collects geographic location data at partial intervals or for partial trip segments rather than continuously throughout entire trips. By applying partial action - collecting location points only when necessary for trip classification (such as at trip start, end, and significant turning points) - the system maintains classification accuracy while significantly reducing the volume of stored location data.
Data Source
AI summary
Implementations include classifying vehicle trips as similar to previous trips based on location information of a vehicle received from a location device. Unique tile identifiers of the trip, each corresponding to a geographic area and the location information, may be determined and used to generate a fingerprint of the trip. The derived trip fingerprint of the trip information may be compared to stored fingerprints of one or more previously received trips to determine if the new trip is similar to one or more of the previous trips. In one instance, information or data of a new trip may be adjusted based on previous trip data. For example, aspects of the new trip or the previous trip may be updated with information or data of a previous trip fi the new trip and the previous trip are similar.