Rider Identification via Motion Sensing and Skeletal Modeling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Ride-hailing services, including autonomous vehicle-based systems, face challenges in accurately identifying intended customers due to difficulties in recognizing subtle human interactions and movements, especially in crowded or obscured environments.
Innovation Solution
The system utilizes sensor data analytics, including video analysis and smart device acceleration, to locate and identify customers by cross-referencing data from smart devices and vehicle or infrastructure sensors, generating skeletal models of human motion and predicting gait patterns, and requesting specific actions from potential customers to enhance confidence in identification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional driver-gesture methods are used for rider identification, then the system is simple to operate, but the identification accuracy deteriorates in crowded or obscured environments
Solution Approach 1:
The system segments the identification process into multiple independent components: video capture, skeletal model generation, gait analysis, and action recognition. Each component processes specific aspects of rider identification separately, improving overall accuracy while maintaining manageable system complexity through modular architecture.
Solution Approach 2:
The system transitions from 2D video frames to 3D skeletal models, adding a dimensional transformation that enables more accurate spatial understanding of rider movements and gestures, thereby improving identification precision without proportionally increasing complexity.
2Measurement precision
If multiple sensor data sources are fused for rider identification, then the measurement precision improves, but the device complexity increases
Solution Approach 1:
The system employs a unified sensor fusion framework that processes multiple data types (video, acceleration, gait data) through a common analytical pipeline. This multi-functional approach allows the same core processing architecture to handle diverse sensor inputs, improving localization accuracy while controlling complexity through standardized handling procedures.
Solution Approach 2:
The system introduces skeletal models as intermediary representations that bridge raw sensor data and final identification decisions. These intermediate models simplify the fusion process by providing a standardized format for comparing multiple data sources, thereby improving accuracy without proportionally increasing system complexity.
3Reliability
If the system requests specific actions from potential customers for verification, then the identification reliability improves, but the time required for identification increases
Solution Approach 1:
The system implements periodic verification through requested actions only when initial identification confidence is insufficient. This conditional periodic verification approach maintains high reliability by confirming uncertain identifications while avoiding unnecessary time delays for clearly identified riders, thus balancing reliability improvement with time efficiency.
Solution Approach 2:
The system uses feedback from action recognition results to iteratively improve identification confidence. By observing whether the requested action matches the predicted action from the skeletal model, the system reliably confirms or rejects identification while minimizing the number of verification steps required, thereby controlling time loss.
Data Source
AI summary
Rider identification systems and methods using motion sensing are disclosed herein. An example method includes obtaining a location of a mobile device associated with an individual, obtaining sensor data from the mobile device that includes a first motion profile, the sensor data also including environment information around the location, generating a motion model for the individual using the environment information, generating a second motion profile using the motion model, comparing the second motion profile to the first motion profile, and confirming when the second motion profile matches the first motion profile to confirm that the individual is at the location.