Ride-Sharing Detection Using Motion Pattern Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing ride-sharing detection technologies are inaccurate and resource-intensive, often relying on location-based tracking and proximity detection methods that suffer from errors and consume significant device and server resources.
Innovation Solution
A method that analyzes motion data from user devices to determine if they are in the same vehicle by comparing the similarity of their motion patterns and proximity, using sensors like IMUs and GPS, and activates sensors only during relevant periods to reduce resource consumption.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If location-based tracking of user devices is used for ride-sharing detection, then the detection can be implemented, but the accuracy is poor due to errors in location determination and timing recording
Solution Approach 1:
The patent replaces location-based tracking (GPS/communication signals) with motion-based detection using IMU sensors. Instead of relying on external positioning systems that suffer from accuracy errors and timing issues, the system uses inertial measurement units to capture motion patterns, which are then compared to detect ride-sharing. This substitution eliminates dependency on location determination accuracy and timing synchronization.
2Measurement precision
If proximity detection by communication signals is used, then nearby user devices can be detected, but false detections occur across adjacent vehicles
Solution Approach 1:
The patent replaces communication signal-based proximity detection with motion pattern comparison using IMU data. Instead of detecting Bluetooth or other communication signals that can be detected across adjacent vehicles causing false positives, the system compares motion patterns captured by inertial sensors. Motion patterns are unique to each vehicle's movement characteristics, providing more reliable ride-sharing detection without false positives from nearby vehicles.
3Measurement precision
If location or proximity data is collected for an entire journey, then ride-sharing detection can be performed, but system resources such as battery, storage, and computation are consumed
Solution Approach 1:
The patent implements periodic sampling of motion data instead of continuous collection. The IMU sensors capture motion patterns at specific intervals or during key journey segments, rather than continuously throughout the entire journey. This periodic approach maintains detection accuracy by capturing sufficient motion characteristics while significantly reducing battery consumption, storage requirements, and computational resources needed for processing.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method of detecting ride-sharing by two or more passengers travelling in a shared vehicle is provided. The method comprises: receiving data indicating that a first device associated with a first of the passengers and a second device associated with a second of the passengers are in physical proximity to each other, receiving first motion data from the first device, receiving second motion data from the second device, and comparing the first motion data and the second motion data to verify if the first and second devices are travelling in a same vehicle.