Autonomous Driving Control System with Dynamic Vigilance Thresholds
Solution Overview
Problem
Existing autonomous driving control systems for vehicles fail to maintain optimal driver vigilance levels, leading to excessive tension or insufficient readiness for manual driving, due to reliance on GPS signals which can be unreliable in certain terrains, without adequately adjusting driver alertness based on the system's output reliability.
Innovation Solution
An autonomous driving control system that includes a driver condition sensor, a reliance calculating part, a vigilance calculating part, and an action control part to adjust the strength of alerts to the driver based on the system's output reliability and driver vigilance, defining sub-regions to maintain suitable vigilance levels regardless of GPS signal quality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the autonomous driving control part uses GPS signals to localize the host vehicle, then the autonomous driving can be performed accurately when GPS signals are received well, but the localization accuracy and driving suitability deteriorate when GPS signal reception is difficult
Solution Approach 1:
The system dynamically adjusts the driver vigilance threshold based on the calculated autonomous driving output reliance. When GPS signals are good and reliance is high, the threshold is lowered allowing relaxed vigilance. When GPS signals deteriorate and reliance drops, the threshold is raised to require higher driver vigilance, adapting to changing conditions in real-time
Solution Approach 2:
The system implements a feedback loop where the reliance calculating part continuously evaluates GPS signal quality and autonomous driving output reliability, feeds this information to the vigilance calculating part, which then adjusts the required driver vigilance threshold accordingly, creating a closed-loop adaptive system
2Reliability
If the predetermined awakeness level is set high, then the driver is ready for manual driving even when autonomous driving output reliance is low, but the driver is maintained in an excessively tense state when reliance is high
Solution Approach 1:
The system replaces the static predetermined awakeness level with a dynamic vigilance threshold that changes based on autonomous driving output reliance. The vigilance calculating part computes appropriate threshold values that adapt to current GPS signal conditions and system reliability, eliminating the need for a fixed high threshold that causes unnecessary tension
3Object-affected harmful factors
If the predetermined awakeness level is set low, then the driver is not excessively tense when autonomous driving output reliance is high, but the driver may not be ready for manual driving when reliance falls
Solution Approach 1:
The system dynamically adjusts the vigilance threshold based on real-time assessment of autonomous driving output reliance. When reliance is high, a lower threshold reduces driver tension. When reliance drops due to poor GPS signals, the threshold automatically increases to ensure driver readiness, replacing the unsafe static low threshold approach
4Ease of operation
If a fixed awakeness threshold is used regardless of autonomous driving output reliance, then the system is simple to operate, but the driver vigilance cannot be suitably maintained across varying GPS signal conditions
Solution Approach 1:
The system transitions from a static fixed threshold to a dynamic adaptive threshold that automatically adjusts based on autonomous driving output reliance calculated from GPS signal quality and other factors, maintaining suitable driver vigilance without requiring complex manual configuration
Solution Approach 2:
The vigilance calculating part automatically determines appropriate driver vigilance thresholds based on system-generated reliability data, eliminating the need for manual threshold setting while ensuring appropriate driver alertness levels are maintained across varying operating conditions
Data Source
AI summary
An autonomous driving control system for a vehicle which is able to switch between manual driving and autonomous driving is provided with a driver condition sensor, acting part, and electronic control unit. The electronic control unit is provided with an autonomous driving control part, reliance calculating part for calculating an autonomous driving output reliance, vigilance calculating part for calculating a driver vigilance, and an action control part for controlling a strength of an action against a driver. In a region in which an operating point determined by the autonomous driving output reliance and driver vigilance can fall, a plurality of sub regions are defined by boundary lines extending so that the driver vigilance becomes higher as the autonomous driving output reliance becomes lower. The action control part controls the strength of the action against the driver to differ in accordance with the sub region in which the operating point falls.