Anomalous Event Detection Using Human Behavior Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems fail to effectively detect and validate anomalous events in a driving environment, such as rubbernecking, which can lead to distracted driving and traffic disruptions, and lack a reliable method to verify crowdsourced information.
Innovation Solution
A method and system that utilize human-machine collaboration to detect inherent human behaviors like rubbernecking by sending inquiries to connected entities spatiotemporally related to a vehicle, uploading driving scene data when a threshold of responses indicates an anomalous event, leveraging both human identification of interesting scenes and machine capture of details.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine-based detection systems are used to detect anomalous events, then detection coverage and data capture capability are improved, but the ability to validate whether an event is truly anomalous deteriorates
Solution Approach 1:
The system uses feedback from human observers to validate and correct machine detection. When the machine detects an anomalous event, it queries human observers to confirm whether the event is indeed anomalous. This feedback loop allows the system to learn from human judgment and improve its validation accuracy over time.
Solution Approach 2:
Human observers serve as an intermediary between the machine detection system and the final validation. The system leverages human intuition and contextual understanding to verify machine detections, combining machine's comprehensive coverage with human's ability to judge anomaly truly.
2Reliability
If crowdsourced information is collected from multiple sources, then detection reliability is improved, but the complexity of verifying and validating the information worsens
Solution Approach 1:
The verification process is segmented into multiple independent queries sent to different human observers. Each observer provides an independent verification, and the system aggregates these results. This segmentation allows complex verification to be broken down into simple, parallel tasks.
Solution Approach 2:
The system sends verification inquiries to a subset of connected entities rather than all possible sources. By querying only relevant connected entities (those spatiotemporally related to the detected event), the system achieves sufficient verification without overwhelming complexity.
3Reliability
If inherent human behavior detection is used to identify anomalous events, then the ability to detect distracted driving improves, but false alarms from normal human reactions worsen
Solution Approach 1:
The system uses feedback from multiple observers to distinguish between genuine anomalous behavior and normal human reactions. When observers consistently report seeing the same behavior pattern, the system confirms it as anomalous. This feedback mechanism reduces false alarms by requiring corroboration.
Solution Approach 2:
The system sets a threshold level of responses required to confirm an anomalous event. By requiring multiple independent observations before triggering an alarm, the system filters out false positives from normal human reactions while maintaining sensitivity to actual anomalies.
Data Source
AI summary
Anomalous events in a driving environment can be detected and/or validated by leveraging inherent human behavior. A notification of the anomalous event can be received from an initial connected entity in the driving environment. In response, an inquiry can be sent to connected entities spatiotemporally related to the initial connected entity as to whether any person associated with the connected entities is exhibiting an inherent human behavior. When a threshold level of responses to the inquiry are received from the connected entities indicating that an inherent human behavior has been detected, the one or more connected entities can be caused to upload driving scene data for at least a time when the inherent human behavior by the occupant of the connected vehicle was detected.