Autonomous Driving Agent Human Trust Calibration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Human trust and cognitive workload in autonomous driving systems are not effectively calibrated, leading to potential misuse due to either under-trust or over-reliance on automation, which can result in unsafe conditions.
Innovation Solution
An autonomous driving agent that uses sensor information to estimate belief states of human trust and cognitive workload, dynamically controlling the display system to provide cues for calibration of automation transparency, thereby influencing human behavior and trust through a Partially Observable Markov Decision Process (POMDP) framework.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the autonomous driving system provides high automation transparency, then human trust is improved, but cognitive workload increases
Solution Approach 1:
The system dynamically adjusts automation transparency based on real-time estimation of human trust and cognitive workload states. The circuitry modifies the level of information provided to the human user dynamically, increasing transparency when trust is low and decreasing it when cognitive workload is high, thereby resolving the contradiction between building trust and maintaining ease of operation
Solution Approach 2:
The system changes the parameter of automation transparency based on estimated belief states of human trust and cognitive workload. By adjusting this parameter dynamically according to the human operator's state, the system optimizes the balance between establishing trust and preventing cognitive overload
2Ease of operation
If the autonomous driving system provides low automation transparency, then cognitive workload is reduced, but human trust decreases
Solution Approach 1:
The system dynamically adjusts automation transparency based on real-time estimation of human trust and cognitive workload states. The circuitry modifies the level of information provided to the human user dynamically, increasing transparency when trust is low and decreasing it when cognitive workload is high, thereby resolving the contradiction between building trust and maintaining ease of operation
Solution Approach 2:
The system uses feedback from observed human behavior (eye tracking, cognitive workload indicators) to adjust automation transparency. By continuously monitoring human responses and adjusting the level of information provided accordingly, the system maintains an optimal balance between building trust and preventing cognitive overload
3Measurement precision
If the system continuously monitors human trust and cognitive workload, then calibration accuracy is improved, but system complexity increases
Solution Approach 1:
The circuitry performs multiple functions using a unified approach: it estimates human trust, estimates cognitive workload, selects automation transparency levels, and controls display outputs all through a single integrated system. This multi-functionality reduces overall system complexity while maintaining high calibration accuracy
Solution Approach 2:
The system estimates belief states for multiple parameters (human trust and cognitive workload) simultaneously and uses these parameter changes to drive a single control decision regarding automation transparency. This coordinated parameter management achieves high calibration accuracy without proportionally increasing system complexity
Data Source
AI summary
An autonomous driving agent is provided. The autonomous driving agent determines a set of observations from sensor information of a sensor system of a vehicle. The set of observations includes human attention information for a scene of surrounding environment and a level of human reliance as indicated by human inputs to the autonomous driving agent. The autonomous driving agent estimates, based on the set of observations, belief states for a first state of human trust on the autonomous driving agent and a second state of human's cognitive workload during journey. The autonomous driving agent selects, based on the estimated belief states, a first value for a first action associated with a level of automation transparency between a human user and the autonomous driving agent and controls a display system based on the selected first value to display a cue for calibration of the human trust on the autonomous driving agent.