Adaptive Trust Calibration in Driving Automation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Human trust and workload dynamics in automated driving systems are not effectively managed, leading to either disuse of automation due to low trust or disengagement from driving processes due to over-trust, which affects the optimal utilization of automation benefits.
Innovation Solution
A computer-implemented method and system that utilize a Markov decision process model to analyze eye gaze direction and driver reliance, processing these factors to determine an optimal level of automation transparency through augmented reality cues and control policies, thereby adjusting autonomous transparency in real-time to match human trust and workload states.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If automation transparency is increased to enhance driver trust, then driver trust improves, but driver workload increases
Solution Approach 1:
The system dynamically adjusts automation transparency based on real-time driver trust and workload states. The Markov decision process model continuously updates the optimal transparency level, transforming the static transparency setting into a dynamic parameter that adapts to changing driver conditions, thereby resolving the contradiction between maintaining high trust and low workload
Solution Approach 2:
The system changes the transparency parameter adaptively based on modeled driver states. By processing eye gaze direction and driver reliance data through the Markov model, the system identifies optimal transparency levels that balance trust enhancement with workload management, effectively managing the trade-off between these two opposing requirements
2Ease of operation
If automation transparency is decreased to reduce driver workload, then driver workload decreases, but driver trust deteriorates
Solution Approach 1:
The system implements feedback loops where driver eye gaze and reliance data are continuously collected and processed. The Markov decision process model uses this feedback to update driver trust and workload states, then adjusts transparency accordingly. This closed-loop feedback mechanism ensures that transparency reductions to lower workload do not excessively compromise trust
Solution Approach 2:
Rather than using a fixed low transparency setting, the system dynamically determines transparency levels based on real-time driver states. This dynamic approach allows the system to reduce transparency only when and where it benefits workload without significantly impacting trust, resolving the contradiction in favor of both reduced workload and maintained trust
3Reliability
If automation transparency is dynamically adjusted based on driver states, then trust calibration improves, but system complexity increases
Solution Approach 1:
The Markov decision process model serves as an intermediary that processes complex driver state data and translates it into transparency adjustment decisions. This intermediary layer manages the complexity by providing a structured framework for analyzing eye gaze and reliance data, thereby improving trust calibration without requiring the entire system to become equally complex
Solution Approach 2:
The system replaces direct mechanical or manual adjustment mechanisms with computational modeling and data processing. By using the Markov model to automatically determine optimal transparency levels based on processed driver data, the system achieves sophisticated trust calibration through software-based mechanisms rather than complex hardware adjustments
Data Source
AI summary
A system and method for providing adaptive trust calibration in driving automation that include receiving image data of a vehicle and vehicle automation data associated with automated of driving of the vehicle. The system and method also include analyzing the image data and vehicle automation data and determining an eye gaze direction of a driver of the vehicle and a driver reliance upon automation of the vehicle and processing a Markov decision process model based on the eye gaze direction and the driver reliance to model effects of human trust and workload on observable variables to determine a control policy to provide an optimal level of automation transparency. The system and method further include controlling autonomous transparency of at least one driving function of the vehicle based on the control policy.