Abnormal Driving Detection Using Gap-Filled Vehicle Trajectories
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving systems face challenges in detecting abnormal driving behavior, particularly when there are gaps in time-series data collection due to lane changes or obstructions, which hinders the ability to predict and respond to abnormal vehicle movements effectively.
Innovation Solution
The system employs a method to fill in data gaps by receiving second driving data from nearby vehicles or remote servers, using machine learning models to generate metadata that helps identify abnormal driving patterns, and implements a model for the ego vehicle to respond to predicted movements, even with incomplete data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional time-series modeling is used to detect abnormal driving behavior, then measurement precision is improved, but reliability deteriorates when data gaps occur due to lane changes or obstructions
Solution Approach 1:
The system performs preliminary actions by collecting and storing driving data from multiple vehicles before gaps occur. Multiple vehicles continuously collect driving data including position, speed, and acceleration, storing this data in advance so that when one vehicle experiences data gaps due to lane changes or obstructions, alternative data sources are already available to maintain detection reliability
Solution Approach 2:
The system introduces an intermediary mechanism by using multiple vehicles as data sources and a central server as a mediator to aggregate and fill gaps in the time-series data. When one vehicle's data is interrupted, the server intermediates by providing data from other vehicles, ensuring continuous and reliable abnormal driving behavior detection without compromising measurement precision
2Reliability
If data collection from multiple vehicles is implemented, then reliability is improved, but device complexity increases
Solution Approach 1:
Each vehicle in the system is designed with multi-functionality, serving both as a data collector for its own abnormal behavior detection and as a data source that can fill gaps for other vehicles. This universal approach allows the system to improve reliability through multiple data sources without requiring complex dedicated infrastructure, as existing vehicle sensors and communication systems are leveraged for multiple purposes
Data Source
AI summary
Systems and methods are provided for programmatically determining an abnormal driving performed by a second vehicle. The systems and methods may, for example, receive first driving data from a sensor of the ego vehicle, the first driving data comprising a gap in time-series data entries of movements of the second vehicle; receive second driving data that fills in the gap in the time-series data entries of the second vehicle; aggregate the first driving data with the second driving data to aggregated driving data to generate metadata, the aggregated driving data comprising movements of the second vehicle that exceed a threshold value of predicted movements defined in a model of abnormal driving detection, and the metadata used to select the model of abnormal driving detection; and execute an algorithm that provides driving assistance for the ego vehicle in responding to the predicted movements defined in the model of abnormal driving detection.