Driver Cognitive Modeling for Autonomous Vehicle Hand-Over Timing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for transferring a motor vehicle from autonomous driving mode to manual driving mode do not effectively consider situational characteristics, leading to suboptimal safety and transition efficiency, as they lack personalized and dynamic adaptation of warning strategies and hand-over processes.
Innovation Solution
A method utilizing a cognitive model of the vehicle driver to simulate perception and decision-making processes, allowing for the estimation of the time required for the driver to take over, which adapts the automated driving function and warning strategy to optimize the hand-over process, considering individual and situational factors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a standardized take-over request is used with predefined conditions, then the system structure remains simple, but the safety and effectiveness of the transfer are insufficient due to lack of situational and individual adaptation
Solution Approach 1:
The cognitive model performs preliminary simulation of the driver's perception and decision-making processes before the actual take-over request is issued. This allows the system to pre-estimate the time required for the driver to take over and adapt the warning strategy accordingly, improving safety without requiring complex real-time adjustments during the transfer process
Solution Approach 2:
A cognitive model is created as a simplified copy of the driver's information processing system. This model replicates the driver's perception processes and decision-making mechanisms, allowing the automated driving system to simulate and predict driver behavior without needing to implement the full complexity of human cognition in the actual vehicle system
2Reliability
If the take-over request is issued immediately, then the response time is short, but the driver may not be sufficiently prepared leading to unsafe take-over
Solution Approach 1:
The system performs preliminary simulation of the driver's information processing time using the cognitive model before issuing the take-over request. This allows the system to determine the optimal timing for the take-over request such that the driver is sufficiently prepared, balancing driver readiness with minimizing the actual take-over time
Solution Approach 2:
The cognitive model provides feedback about the estimated time required for the driver to process information and execute the take-over action. This feedback is used to adapt the warning strategy and timing of the take-over request, ensuring the driver is ready without unnecessarily extending the take-over process
3Measurement precision
If the cognitive model simulates detailed perception and decision-making processes, then the accuracy of take-over time estimation improves, but the computational complexity increases
Solution Approach 1:
The cognitive model creates a simplified copy of the driver's information processing system that captures the essential perception and decision-making processes. This model replicates key cognitive mechanisms without implementing the full complexity of human cognition, achieving sufficient estimation accuracy while maintaining computational efficiency
Solution Approach 2:
The cognitive model uses parameterized representations of perception and decision-making processes that can be adjusted based on the specific driving situation and driver characteristics. This allows the system to achieve accurate take-over time estimation by adjusting model parameters rather than implementing complex detailed simulations of all cognitive processes
Data Source
AI summary
A method for transferring a motor vehicle from an autonomous driving mode, in which the motor vehicle is guided autonomously, into a manual driving mode, in which the motor vehicle is guided by a vehicle driver. In the method, pieces of information for supporting the transfer are ascertained with the aid of a cognitive model of the vehicle driver, the cognitive model describing at least one perception process of the vehicle driver with respect to a driving situation and at least one decision-making process of the driver with respect to an action option. A device configured for executing the method is also described.