Joint Motion Detection Using Multi-Sensor Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Service providers face challenges in accurately detecting whether two or more people are traveling together in the same transportation vehicle, as existing methods relying on location data alone can result in false positives or negatives, especially in areas with poor GPS reception or traffic jams, and cannot distinguish between true joint motion and nearby individuals.
Innovation Solution
A system that uses multiple sensor types, such as GPS, IMU, barometers, Bluetooth, NFC, and acoustics, to collect and combine data for more accurate joint motion prediction, employing classifiers optimized for each sensor type to integrate sensor-specific predictions into a unified joint motion prediction, allowing for detection even without location data and distinguishing between true joint motion and separate individuals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If location data alone is used for joint motion detection, then the system is simple to implement, but the detection accuracy deteriorates due to false positives and negatives
Solution Approach 1:
The patent combines multiple sensor types (GPS, IMU, barometer, Bluetooth, NFC, acoustics) to detect joint motion, merging their respective strengths to overcome the limitations of any single sensor. This fusion approach improves detection accuracy by cross-validating signals and reducing false positives/negatives that occur when using location data alone.
2Measurement precision
If multiple sensor types are used for joint motion detection, then the detection accuracy improves, but the device complexity increases
Solution Approach 1:
The patent segments the joint motion detection system into independent sensor modules, each handling its own data processing and classification. Each sensor type (GPS, IMU, barometer, etc.) has its dedicated classifier that processes data independently, then results are combined. This modular segmentation reduces overall system complexity by distributing processing tasks rather than requiring one complex centralized system.
Solution Approach 2:
The patent implements a universal classification framework that handles multiple sensor types through a common architecture. The same basic classification and fusion logic applies regardless of which sensor type is being processed, making the system multi-functional while maintaining manageable complexity through code reusability and standardized interfaces.
3Loss of information
If GPS data is used in poor reception areas, then location information is available, but the reliability of joint motion detection deteriorates
Solution Approach 1:
The patent introduces intermediary sensors (IMU, barometer, Bluetooth, NFC, acoustics) that can detect joint motion through alternative physical principles not dependent on GPS signals. These intermediary sensors act as mediators when GPS reception is poor, providing backup detection mechanisms that maintain reliability by detecting correlated motion patterns, proximity, or vehicle presence through other means.
Data Source
AI summary
An approach is provided for performing ride-sharing functions based on joint motion using multiple sensor data. The approach, for example, involves retrieving a joint motion prediction indicating whether at least two devices are traveling in a same transportation vehicle. The joint motion prediction is computed based on sensor data collected from the at least two devices using at least one sensor type from among a plurality of sensor types. Each sensor type of the plurality of sensor types is associated with a respective joint motion classifier configured to compute a sensor-type joint motion prediction that is used for generating the joint motion prediction. The approach also involves initiating a ride-sharing function for respective users of the at least two devices based on the joint motion prediction.