Abnormal Driving Origin Estimation Using Happens-Before Analysis

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing driver monitoring systems (DMS) generate false positives when identifying the source of abnormal driving, often erroneously attributing erratic vehicle behavior to the follower vehicle instead of the subject vehicle due to fluctuations in separation distance data.

Innovation Solution

A detection system that employs parallel computations and happens-before analysis to accurately determine the origin of abnormal driving by synchronizing and ordering abnormal classifications from multiple vehicle trajectories, reducing false positives through time-series analysis and machine learning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a driver monitoring system uses sensor data to detect abnormal driving behavior, then it can identify potential safety issues, but it generates false positives by erroneously attributing erratic behavior to the follower vehicle instead of the subject vehicle

Engineering Contradiction:

Improveaccuracy of abnormal driving detectionVSAvoidprecision in identifying the origin of abnormal behavior

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The system segments the analysis by separately evaluating the subject vehicle and follower vehicle independently through parallel computations. Each vehicle's abnormal classification is determined separately based on its own driving patterns, rather than analyzing them as a single coupled system. This segmentation allows the system to trace the origin of abnormal behavior more accurately by comparing independent classifications.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system performs preliminary abnormal classification for both the subject vehicle and follower vehicle before determining the origin of abnormal behavior. By pre-computing abnormal classifications for each vehicle based on their respective driving patterns, the system establishes a foundation for accurate origin identification through happens-before analysis, preventing false positives.

Inventive Principle:

Principle #10Preliminary action

2Difficulty of detecting and measuring

If the system analyzes fluctuations in separation distance between vehicles, then it can detect potential distracted driving, but it misidentifies the source of abnormal behavior due to aggressive driving by the subject vehicle

Engineering Contradiction:

Improveease of detecting abnormal behaviorVSAvoidaccuracy in locating the source of abnormal behavior

Core Design Contradiction:

Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The detection process is segmented into independent analysis streams for the subject vehicle and follower vehicle. Each vehicle's behavior is evaluated separately using its own driving patterns, allowing the system to easily detect abnormal behavior while accurately attributing it to the correct vehicle through comparative analysis of segmented results.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

Instead of assuming the follower vehicle is the source of abnormal behavior (conventional approach), the system inverts the analysis by independently evaluating both vehicles and using happens-before analysis to determine which vehicle actually initiated the abnormal behavior. This inversion prevents misattribution caused by analyzing only the follower vehicle's perspective.

Inventive Principle:

Principle #13The other way round (Inversion)

3Adaptability or versatility

If the system uses parallel computations to classify abnormal behavior for multiple vehicles, then it improves detection coverage, but it requires complex synchronization and ordering of classifications

Engineering Contradiction:

Improvecapability to analyze multiple vehicle trajectoriesVSAvoidcomplexity of synchronizing and ordering classifications

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs preliminary abnormal classification for each vehicle trajectory in parallel before synchronization is needed. By pre-computing abnormal classifications for all vehicles independently, the system reduces the complexity of subsequent synchronization, as each vehicle already has its classification ready for straightforward temporal ordering through happens-before analysis.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The happens-before analysis acts as an intermediary mechanism that simplifies the synchronization of parallel computations. Instead of implementing complex inter-process communication and locking mechanisms, the system uses logical temporal relationships (happens-before) to order classifications, providing a simpler coordination approach for parallel vehicle trajectory analysis.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS12358536B2Systems and methods for estimating the origins of abnormal driving

Publication Date: 2025.07.15 TOYOTA MOTOR ENG & MFG NORTH AMERICA INC

AI summary

Systems, methods, and other embodiments described herein relate to estimating the origins of abnormal driving through observed driving patterns. In one embodiment, a method includes detecting, using parallel computations, abnormal classifications associated with a subject vehicle and nearby vehicles according to driving patterns derived from observation data, the abnormal classifications being associated with exceeding a position range in the observation data. The method also includes estimating an origin of abnormal driving with happens-before analysis according to the abnormal classifications, and the abnormal driving is associated with deviations in a traffic flow associated with the subject vehicle and the nearby vehicles. The method also includes controlling the subject vehicle using a driving command according to the origin.