Distracted Driving Event Clustering for Driver-Passenger Phone Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems inaccurately distinguish between distracted driving events involving the driver and passenger phone usage, leading to unreliable risk assessments for personal injury and property loss.
Innovation Solution
A distracted driving analysis system employing unsupervised, semi-supervised, and supervised machine learning techniques to cluster and label driving and phone usage data, using features like sudden braking and phone interactions to differentiate between driver and passenger events.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If current systems monitor phone usage to detect distracted driving, then distracted driving detection capability is provided, but accuracy in distinguishing driver phone usage from passenger phone usage deteriorates
Solution Approach 1:
The system segments phone usage events into distinct categories by analyzing multiple features simultaneously. It divides driving events into clusters (e.g., distracted driving vs. passenger usage) by segmenting the data based on phone interaction patterns, vehicle dynamics, and contextual features, thereby improving classification accuracy
Solution Approach 2:
The system applies different analysis methods to different aspects of phone usage data. It examines local characteristics such as phone interaction timing, duration, intensity, and correlation with specific driving events to accurately determine whether phone usage occurred during distracted driving or passenger activities
2Productivity
If phone usage monitoring is used to assess driver risk, then risk assessment functionality is provided, but reliability of risk quantification deteriorates due to inaccurate identification
Solution Approach 1:
The system incorporates feedback mechanisms where classification results are continuously refined. It uses labeled training data to train machine learning models, and the models provide feedback by improving their classification accuracy over time, thereby enhancing the reliability of risk quantification
Solution Approach 2:
The system replaces simple rule-based detection mechanisms with advanced machine learning algorithms. It substitutes traditional mechanical monitoring approaches with intelligent systems that can learn from data patterns, improving the reliability of risk assessment through semi-supervised and supervised learning techniques
Data Source
AI summary
A distracted driving analysis system for identifying distracted driving events is provided. The system includes a processor in communication with a memory device programmed to: (i) receive driving event records, each driving event record including phone usage by a user, wherein a driving event record is labeled as an actual distracted driving event or a passenger event, (ii) divide the driving event records into at least two clusters based at least in part upon common features and the labels of each driving event record by processing the plurality of driving event records with a semi-supervised machine learning algorithm, (iii) generate a trained model based at least in part upon the at least two clusters, (iv) process a new driving event using the trained model, (v) assign the new driving event to one of the clusters using the trained model, and/or (vi) determine whether the new driving event is an actual distracted driving event or a passenger event.