Driver Dozing-Based Vehicle Deceleration for Secondary Collision Avoidance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle driving assist systems fail to effectively prevent secondary collisions without causing unnecessary deceleration of the vehicle, as they either wait until the driver's dozing state is confirmed or apply excessive deceleration when the driver is only partially dozing.
Innovation Solution
A vehicle driving assist apparatus with an electronic control unit that assesses both collision and dozing levels to execute a forcibly-decelerating process, adjusting the deceleration based on the driver's dozing level, using either a driving force limiting process or a braking process, to avoid secondary collisions while minimizing unnecessary deceleration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If the forcibly-decelerating process is executed only when the driver is dozing with certainty (determined after a certain time), then unnecessary deceleration is avoided, but the secondary collision may not be avoided in time
Solution Approach 1:
The patent applies dynamics by making the deceleration control adaptive and variable based on the dozing level detection results. The ECU dynamically adjusts whether to execute forcibly-decelerating control according to the detected dozing level, transitioning from a static fixed-threshold approach to a dynamic multi-level response system that optimizes both safety and energy efficiency.
Solution Approach 2:
The patent changes the parameter of dozing level detection from a binary certainty-based determination to a multi-level assessment system. By detecting dozing level as a continuous or graded parameter rather than a fixed threshold, the system can differentiate between light dozing and deep dozing states, enabling appropriate control responses for each level.
2Reliability
If the forcibly-decelerating process is executed also when the driver is dozing with low certainty, then the secondary collision can be avoided, but the own vehicle may be unnecessarily decelerated
Solution Approach 1:
The patent segments the dozing state into multiple levels (low certainty dozing and high certainty dozing) based on the duration and characteristics of eye closure detection. This segmentation allows the system to apply different control strategies for different dozing levels, avoiding unnecessary deceleration for light dozing while ensuring safety for deep dozing states.
Solution Approach 2:
The patent applies partial action by executing forcibly-decelerating control only for specific dozing levels rather than all dozing states. The ECU selectively applies deceleration based on the detected dozing level, performing the action partially (only when necessary) rather than excessively (for all dozing cases), thus balancing safety and energy efficiency.
3Measurement precision
If the dozing determination condition continues for a certain time before execution, then the driver state is confirmed with certainty, but the response time to avoid secondary collision is delayed
Solution Approach 1:
The patent applies preliminary action by continuously monitoring and detecting the driver's eye closure state in advance, accumulating detection data to determine dozing level before a collision occurs. The ECU performs preliminary dozing level assessment and is ready to execute deceleration control immediately when the assessment confirms deep dozing, reducing response delay while maintaining accuracy.
Solution Approach 2:
The patent implements feedback by continuously detecting the driver's eye closure state and using this feedback to update the dozing level determination. The ECU uses real-time detection feedback to adjust the control strategy, ensuring accurate dozing determination while maintaining rapid response capability through continuous monitoring rather than waiting for fixed time intervals.
Data Source
AI summary
A vehicle driving assist apparatus acquires a collision index value which represents a magnitude of a collision of an own vehicle and a dozing level of a driver of the own vehicle, and executes a secondary collision reducing control of executing a forcibly-decelerating process of forcibly decelerating the own vehicle when a light collision condition is satisfied, and a dozing condition is satisfied. The vehicle driving assist apparatus executes the forcibly-decelerating process so as to decelerate the own vehicle with controlling a deceleration of the own vehicle such that the deceleration of the vehicle realized when the light collision condition and the dozing condition become satisfied, and the dozing level is relatively low, is smaller than the deceleration of the own vehicle realized when the deceleration when the light collision condition and the dozing condition become satisfied, and the dozing level is relatively high.