Driver Gaze-Based Mode Transition Control in Automated Vehicles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle control systems lack the ability to effectively transition between autonomous and manual driving modes based on the occupant's engagement with non-driving tasks, as they do not account for individual differences in attention levels and task types, leading to inadequate transition profiles.
Innovation Solution
A control system that utilizes image data and gaze characteristics to determine an occupant's attention level, adjusting transition profiles and alerts to ensure seamless mode transitions by quantifying gaze direction, duration, saccade velocity, and saccade length, and integrating this data with physiological sensor information to tailor the transition process to the occupant's engagement level.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a fixed transition profile is used for mode transitions, then the control system is simple to implement, but it cannot adapt to individual differences in occupant attention levels and task types
Solution Approach 1:
The system performs preliminary assessment of the occupant's attention level by analyzing gaze characteristics (direction, duration, saccade velocity, saccade length) before initiating the mode transition. This preliminary action allows the system to pre-determine the appropriate transition profile based on the occupant's current engagement state, ensuring the transition is tailored to individual needs without adding complex real-time adjustments during the transition itself.
Solution Approach 2:
The system changes the parameters of the transition profile (such as transition duration, alert intensity, and warning timing) based on the quantified gaze characteristics. By dynamically adjusting these parameters according to the occupant's attention level and the specific non-driving task being performed, the system achieves adaptability while maintaining a structured approach to transition management.
2Reliability
If the system provides detailed alerts and extended transition profiles for occupants engaged in non-driving tasks, then safety is improved, but occupant annoyance increases
Solution Approach 1:
The system applies partial action by providing only the necessary level of alerting and transition support based on the occupant's actual engagement with the non-driving task. Rather than always providing maximum alerts, the system quantifies gaze characteristics to determine the appropriate degree of intervention, providing enhanced alerts only when the occupant's attention is sufficiently compromised, thereby maintaining safety while reducing unnecessary annoyance.
Solution Approach 2:
The system uses feedback from gaze analysis to continuously monitor the occupant's attention state during the transition process. By analyzing changes in gaze characteristics in response to alerts, the system can adjust the transition profile in real-time, reducing alert intensity or extending transition duration only as much as necessary to ensure safe mode switching, thus balancing safety requirements with occupant comfort.
3Productivity
If the system quickly transitions from autonomous to manual mode, then productivity is improved, but the occupant may not be ready to take control, reducing safety
Solution Approach 1:
The system performs preliminary assessment of occupant readiness by analyzing gaze characteristics before initiating the mode transition. By determining the occupant's attention level and engagement state in advance, the system can predict the appropriate transition duration and alert sequence needed to ensure the occupant is ready to take control, thereby maintaining both speed and safety in the transition process.
Solution Approach 2:
The system dynamically adjusts the transition profile parameters (such as transition duration, alert timing, and warning intensity) based on the quantified gaze characteristics. This dynamic adjustment allows the transition speed to be optimized for each individual situation, enabling faster transitions when the occupant is already attentive while providing extended, more gradual transitions when the occupant is deeply engaged in a non-driving task, thus balancing productivity and safety.
Data Source
AI summary
Aspects of the present disclosure relate to a control system, system and method for controlling a transition between a first driving mode and a second driving mode of a vehicle. The present disclosure relates to receiving image data of an occupant of the vehicle; quantifying one or more characteristics of the occupant's gaze based at least in part on the received image data, the one or more characteristics being indicative of the occupant's attention to a non-driving task; determining a transition profile for transitioning between the first driving mode and the second driving mode based at least in part on the one or more characteristics of the occupant's gaze; and generating and outputting a control signal to transition from the first driving mode to the second driving mode, the control signal comprising instructions for controlling one or more vehicle systems in accordance with the determined transition profile.