Vehicle Control Transfer Policy for Low-Distraction Autonomy Handoffs
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing semi-autonomous vehicle control systems require significant user interaction for transferring control between manual and autonomous modes, which can divert user attention and increase cognitive load, leading to potential distractions and safety concerns.
Innovation Solution
A system that uses an input module to acquire user preferences and generate a transfer of control (TOC) policy based on current context, allowing automatic transitions between control states using a Markov decision process, while estimating cognitive and sensory loads to minimize user interaction and maximize user satisfaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If user interactions are required for transferring control between manual and autonomous modes, then control transfer reliability is improved, but user attention is diverted and cognitive load increases
Solution Approach 1:
The system performs preliminary actions by continuously monitoring context parameters (vehicle speed, acceleration, steering angle, sensor data) and pre-evaluating transfer conditions before actual control transfer occurs. This allows the system to prepare for control transfer in advance, ensuring reliability while minimizing the need for active user interaction at the moment of transfer.
Solution Approach 2:
The system implements feedback mechanisms by continuously monitoring user state (via cameras, microphones, biometric sensors) and vehicle context, then adjusting control transfer timing and notifications accordingly. This feedback loop ensures reliable control transfer while adapting to user attention levels and cognitive load, reducing unnecessary interactions.
2Ease of operation
If automatic control transitions are implemented, then user interaction is reduced and safety is enhanced, but adaptability to user preferences and context is worsened
Solution Approach 1:
The system performs preliminary learning of user preferences during initial operation phases, storing context-state preferences that guide automatic control transitions. This preliminary action enables the system to adapt to individual user preferences without requiring real-time interactions, maintaining both safety and personalization.
Solution Approach 2:
The system changes operational parameters by dynamically adjusting control transfer thresholds and notification methods based on learned user preferences and current context. This allows automatic transitions to adapt to different users and situations while maintaining reduced interaction levels.
3Loss of information
If control transfer notifications are presented to users, then user awareness is improved, but cognitive load and distraction increase
Solution Approach 1:
The system applies local quality by presenting notifications with characteristics tailored to the specific context and user state. For example, notification modality (visual, auditory, haptic), timing, and intensity are locally optimized based on current driving conditions and user attention levels, ensuring awareness without excessive cognitive load.
Solution Approach 2:
The system changes notification parameters dynamically based on context and learned user preferences. Notification timing, modality, and intensity are adjusted as parameters to minimize cognitive load while maintaining adequate user awareness of control state transitions.
Data Source
AI summary
A system for facilitating transfers of control includes an input module configured to acquire a set of user preferences relating control states to contexts, the control states including at least a manual control state, and an autonomous and/or semi-autonomous control state. The system includes a processing device configured to automatically perform, during operation of a semiautonomous system, generating a transfer of control (TOC) policy based on the set of user preferences and a current context state, the TOC policy prescribing transitions between control states in response to actions. The processing device is also configured to perform, based on an action performed by the user or a control system, determining whether to perform a transition from a current control state to a second control state, and in response to the TOC policy prescribing the transition, transitioning from the current control state to the second control state.