Automated Driving Takeover Notification via Multi-Modal Driver Activity Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automated driving systems for motor vehicles face challenges in reliably communicating the need for driver takeover, particularly when the driver is engaged in secondary activities, which can lead to missed or misinterpreted requests, affecting safety and acceptance.
Innovation Solution
The method employs a vehicle guidance system that detects the driver's activity type and uses multiple information output modalities (audible, optical, haptic) to progressively direct the takeover request, initially undirected and then directed at the driver, ensuring multi-sensory engagement and attention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a single information output modality is used to notify the driver of takeover requests, then the system complexity is reduced, but the reliability of driver notification deteriorates when the driver is engaged in secondary activities
Solution Approach 1:
The notification system is segmented into multiple independent output modalities (acoustic, optical, haptic) that can be activated separately or in combination. Each modality operates through different perception channels, allowing the system to selectively engage the driver through the most effective channel based on the detected activity type, thereby improving notification reliability without requiring all modalities to operate simultaneously
Solution Approach 2:
The system dynamically adapts the information output modality based on real-time detection of driver activity types. The controller selects from multiple notification modalities (acoustic, optical, haptic) depending on what the driver is currently doing, making the notification system flexible and context-aware. This dynamic adaptation ensures reliable notification while maintaining system simplicity by only activating the necessary modality in each situation
2Ease of operation
If the driver is engaged in secondary activities during automated driving, then the driver's attention to takeover requests decreases, but the ability to perform secondary activities improves driver comfort and productivity
Solution Approach 1:
The system continuously monitors and detects the driver's current activity type (such as reading, using mobile devices, or other secondary tasks) and uses this feedback information to adapt the notification modality. The controller receives feedback about driver engagement level and activity state, then selects the appropriate notification channel to ensure the takeover request is effectively communicated despite the driver's secondary activities
Solution Approach 2:
The system changes the parameters of information output by selecting different modalities (acoustic, optical, haptic) based on the detected activity type. When the driver is engaged in visual secondary activities, the system may switch to acoustic or haptic notifications; when the driver is engaged in auditory activities, optical or haptic modalities are selected. This parameter change in notification delivery ensures reliable communication while allowing driver comfort
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for the automated driving of a motor vehicle 1 occupied by a driver 3, wherein during the automated driving of the motor vehicle 1 a request to take over the driving task is made to the driver 3.For reliable transmission of the driving task takeover request to the driver 3, the invention proposes that an activity type of an activity performed by the driver 3 is detected, that depending on the detected activity type, a first information output modality for issuing the driving task takeover request is selected, that the driving task takeover request to the driver 3 is made in the selected first information output modality and is spatially undirected, that subsequently the driving task takeover request is made in the selected first information output modality and is spatially directed towards the driver 3, and that simultaneously the driving task takeover request to the driver 3 is made in a second information output modality different from the first information output modality.