Driver Assistance Behavior Modeling for Aberrant Driving Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing onboard advanced driver assistance systems (ADAS) are unable to provide a robust evaluation of repeated driver behaviors over meaningful time scales and do not effectively integrate this evaluation into route planning or other driving operations.
Innovation Solution
An onboard driver assistance system evaluates driver operations over time scales of tens of seconds to hours, identifying aberrant behaviors such as lane departures, delayed braking, and suboptimal maneuvers by comparing them against a driver behavior model, and takes corrective actions or alerts the driver.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing ADAS operate according to instantaneous vehicle status over a short time scale, then the system response speed is fast, but the system cannot provide a robust evaluation of repeated behaviors
Solution Approach 1:
The system pre-establishes a behavior model containing expected driver behaviors and evaluation criteria before actual driving evaluation. This preliminary model enables the system to robustly evaluate repeated behaviors by comparing actual driver actions against pre-defined expectations, resolving the contradiction between evaluation reliability and time scale.
Solution Approach 2:
The system implements continuous feedback by comparing actual driver behavior against the pre-established behavior model over extended time scales. This feedback mechanism allows the system to accumulate evidence of repeated behaviors and provide robust evaluations, while still maintaining responsive intervention when aberrant patterns are detected.
2Reliability
If the system evaluates driver behavior over extended time scales, then the evaluation robustness improves, but the system cannot provide immediate corrective action
Solution Approach 1:
The system dynamically adjusts its evaluation approach by operating on multiple time scales simultaneously. It maintains continuous monitoring for immediate corrective actions on short time scales, while parallel long-term evaluation tracks repeated behaviors for robust pattern recognition. This dynamic multi-scale approach resolves the contradiction between evaluation robustness and response speed.
Solution Approach 2:
By pre-establishing the behavior model with expected behaviors and evaluation criteria, the system enables immediate comparison of actual driver actions against expectations. This preliminary preparation allows the system to provide immediate corrective feedback when aberrant behaviors are detected, while the extended time scale evaluation continues in parallel to build robust patterns.
3Adaptability or versatility
If the system tracks multiple driver behaviors including suboptimal maneuvers, then the evaluation comprehensiveness improves, but the system complexity increases
Solution Approach 1:
The system segments driver behavior evaluation into distinct categories including lane departures, delayed braking, failure to obey traffic signals, and suboptimal maneuvers. Each category is evaluated against specific criteria in the pre-established behavior model. This segmentation allows comprehensive evaluation of multiple behavior types while managing system complexity through structured, modular assessment.
Solution Approach 2:
The behavior model serves as a universal framework that handles multiple types of driver behaviors through a common evaluation structure. This multi-functional model can assess various aberrant behaviors (lane departures, braking issues, signal obedience, suboptimal maneuvers) using unified principles, thereby improving evaluation comprehensiveness without proportionally increasing system complexity.
Data Source
Figure 1A
Figure 1B
Figure 1C~1D
AI summary
The technology relates to identifying and addressing aberrant driver behavior. Various driving operations may be evaluated over different time scales and driving distances (902, 906). The system can detect driving errors and suboptimal maneuvering, which are evaluated by an onboard driver assistance system and compared against a model of expected driver behavior (908). The result of this comparison can be used to alert the driver or take immediate corrective driving action (912, 914). It may also be used for real-time or offline training or sensor calibration purposes (912, 914). The behavior model may be driver-specific, or may be a nominal driver model based on aggregated information from many drivers. These approaches can be employed with drivers of passenger vehicles, busses, cargo trucks and other vehicles.