Automated Driving Takeover Prediction for Driver Trust Retention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In partially automated vehicles, frequent manual takeovers by drivers due to lack of trust or uncertainty negatively impact traffic safety and energy balance, reducing the effectiveness of automated driving.
Innovation Solution
A method and device using a trained neural network model to predict driver intentions for manual takeovers, providing informed and personalized feedback to drivers to maintain automated driving, adjusting vehicle behavior, and sharing data across a fleet to optimize automated driving conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If drivers manually take over vehicle control due to lack of trust or uncertainty, then drivers maintain control of the vehicle, but traffic safety and energy balance deteriorate
Solution Approach 1:
The system continuously monitors driving conditions and provides feedback to the driver about the automated system's performance and confidence level. When the automated driving function detects improved conditions or high confidence in its capabilities, it communicates this to the driver, encouraging them to maintain automated mode rather than taking over manually.
Solution Approach 2:
The system predicts potential manual takeover intentions before they occur by analyzing driver behavior patterns and driving conditions. By anticipating when a driver might take over, the system can proactively communicate its reliability and adjust its operation to prevent unnecessary takeovers, thereby maintaining automated driving longer and improving traffic safety and energy balance.
2Productivity
If drivers manually take over vehicle control, then drivers assume direct control, but traffic flow and energy balance are negatively affected
Solution Approach 1:
The system provides continuous feedback to drivers about the benefits of maintaining automated driving, including information about traffic flow efficiency and energy consumption. This feedback loop helps drivers understand that staying in automated mode contributes to better overall traffic flow and energy balance, reducing unnecessary manual takeovers.
Solution Approach 2:
The automated driving system monitors and adjusts its own operation based on traffic conditions and driver behavior patterns. It autonomously determines when to communicate with the driver about its reliability and when to maintain automated control, thereby self-optimizing its operation to maximize traffic flow efficiency and energy balance while minimizing disruptive manual takeovers.
3Measurement precision
If a model predicts manual takeover with high accuracy, then driver behavior can be anticipated, but driver trust may be reduced if predictions are too accurate
Solution Approach 1:
The system applies a probability threshold to its predictions, only acting on predictions that exceed a certain confidence level. This partial action approach allows the system to benefit from accurate predictions while avoiding over-reacting to low-confidence predictions that might alienate drivers. The threshold is tuned to balance prediction accuracy with maintaining driver trust.
Data Source
AI summary
A method for increasing the share of automated driving in an at least partially automated vehicle involves monitoring the conditions for automated driving. The number of manual takeovers by the driver is reduced in order to utilize the advantages of automated driving optimally designed for traffic flow, traffic safety and energy balance, when conditions for automated driving are met, it is predicted by a model that automated driving will be ended by manual vehicle control. In the event that a takeover by the driver is predicted, information is output to the driver which informs the driver that automated driving is reliably in control of the driving situation.