Project Objectives: Based on consequence-based privilege and confidence-based human sensing, the privileged sensing framework aims to incorporate insights into operator state, capabilities, and intention to optimally fuse inputs from both human and autonomy to provide better decisions in human-autonomy systems.
- Bayesian Human Information Fusion
Abstract: The privileged sensing framework is built upon the concepts of privilege and confidence.
- Privilege is used to define a rigorous mechanism to dynamically arbitrate between human and autonomy decisions through assessing consequences and leveraging operator knowledge, which enables the system to treat the operator inputs as sophisticated, yet noisy sensed information with special rights that are not afforded to other sensor data.
- Confidence provides a measure of reliability in all sensed or estimated signals, thereby enabling optimal integration of information.
- Combining the concepts of privilege and confidence will create a collaborative and generalized framework that enables the integration of the human adaptive abilities for complex and dynamic conditions with autonomy that compensates for human limitations.
Currently, I am collaborating with researchers from AFRL and ARL to develop a Bayesian control architecture to incorporate the concepts of privilege and confidence to effectively integrate human and autonomy decisions.