Driver Behavior Detection Using Moving-Average Pattern Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current technologies lack the ability to accurately detect driver behavior and identify patterns and changes in driving habits, necessitating an improved mechanism for driver behavior detection in computing systems.
Innovation Solution
A computing system that generates a filtered data set based on driving behavior, groups it by origin, creates labels, and compares operating patterns to detection averages to identify changes, sending notifications for display, using deviation speed profiles or acceleration speed profiles to calculate proximity scores and determine threshold exceedance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If current driver behavior detection technologies are used, then basic driving data can be collected, but accurate detection of driver behavior patterns and changes cannot be achieved
Solution Approach 1:
The patent segments driver behavior detection into multiple dimensions: speed-based segmentation (comparing actual speed with speed limits to generate speed profiles), acceleration-based segmentation (generating acceleration profiles from force data), and temporal segmentation (comparing consecutive trips to detect behavior changes). This multi-dimensional segmentation enables accurate detection of driving patterns without requiring overly complex hardware systems.
Solution Approach 2:
The system performs preliminary actions by collecting and storing driving data (speed, acceleration, location) during normal operation, then processing this data offline to generate speed profiles, acceleration profiles, and proximity scores. This preliminary data collection and processing approach allows accurate behavior detection while keeping the real-time system relatively simple.
2Measurement precision
If detailed driving behavior analysis is implemented, then accurate pattern identification is achieved, but computational resources and processing time increase
Solution Approach 1:
The patent applies partial action by focusing computational resources on key behavioral indicators rather than analyzing all possible driving parameters. The system calculates proximity scores based on speed deviations and acceleration patterns, which are the most significant indicators of behavior changes. This selective approach achieves accurate pattern identification while reducing overall computational burden and processing time.
Solution Approach 2:
The system performs preliminary filtering and organization of driving data during normal operation, structuring speed and acceleration data into comparable profiles before detailed analysis is needed. This preliminary structuring reduces the computational complexity of subsequent behavior pattern analysis.
3Speed
If real-time driver behavior monitoring is implemented, then immediate detection of behavior changes is possible, but system complexity and resource consumption increase
Solution Approach 1:
The system performs preliminary actions by continuously collecting and organizing driving data (speed, acceleration, location) during normal operation, preparing it in advance for rapid comparison. When behavior change detection is needed, the system can quickly compare current trip data against previously stored profiles without requiring complex real-time computation, thus achieving fast detection with relatively simple system architecture.
Solution Approach 2:
The patent uses copying by creating profiles of normal driving behavior patterns from historical data, then comparing current driving behavior against these stored profiles. This copy-and-compare approach enables rapid real-time detection of behavior changes without requiring complex analytical systems, as the heavy lifting of pattern recognition was already done when creating the reference profiles.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computing system and method for operating the computing system including: a control unit configured to: generate a filtered data set based on a data set; group the filtered data set based on a previous origin indicator to obtain a device group data; generate a label based on the device group data; generate a training moving average based on the label and the device group data; generate an operating pattern based on the training moving average; identify a change in a driving behavior based on comparing the operating pattern and a detection moving average; generate a notification based on the change; and a communication unit, coupled to the control unit, configured to: send the notification for display on a display interface.