Adaptive Supervisory Control for Human-Cyber-Physical Workload
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current human supervised large-scale cyber-physical systems lack an integrated process for understanding and adapting human supervisory control issues, leading to operator overload due to excessive data, with no effective method to allocate attention to the most relevant events in a timely manner.
Innovation Solution
An integrated system that measures and assesses operator attention and workload using psycho-physiological sensors and human performance models, generating context-sensitive recommendations to optimize attention allocation and maintain acceptable workload levels, while enabling a model-based system to learn from operator actions and adapt the user interface accordingly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a human supervisor oversees a large-scale sensor network with thousands of detectors and hundreds of cameras, then comprehensive monitoring coverage is achieved, but the human supervisor becomes overwhelmed by the large amounts of data streaming in real time
Solution Approach 1:
The patent introduces an intermediary system consisting of computational elements and adaptive user interface that mediates between the large-scale sensor network and the human supervisor. This intermediary processes, filters, and prioritizes data from thousands of sensors and hundreds of cameras, presenting only relevant information to the operator, thereby maintaining comprehensive monitoring coverage while preventing operator overload
Solution Approach 2:
The patent replaces the mechanical approach of direct human supervision of all sensors with an automated computational system. The adaptive user interface and contextual event detection algorithms automatically process and prioritize data streams, substituting manual data filtering with automated computational methods, thus reducing operator workload while maintaining monitoring effectiveness
2Loss of information
If all data from sensors and cameras are presented to the supervisor in real time, then complete information availability is achieved, but the supervisor cannot allocate attention to the most relevant events in a timely fashion
Solution Approach 1:
The patent implements preliminary action by pre-processing data from sensors and cameras through automated detection algorithms that identify contextual events before presenting them to the supervisor. The system anticipates which events will be relevant and prepares prioritized information in advance, allowing the supervisor to focus attention on pre-identified critical events rather than searching through all data streams
Solution Approach 2:
The patent applies local quality by providing different levels of information detail to different aspects of the monitoring task. The adaptive user interface presents summarized, high-level information for most sensors while providing detailed, real-time data only for contextual events that require supervisor attention, thereby maintaining information availability while enabling timely attention allocation
3Loss of information
If the user interface presents all available information to the operator, then complete situational awareness is achieved, but the operator becomes overwhelmed and performance deteriorates
Solution Approach 1:
The patent applies partial action by selectively presenting only the necessary portion of available information to the operator at any given time. The adaptive user interface determines which sensors and events require operator attention and presents information in a prioritized manner, providing complete situational awareness for critical events while omitting or summarizing less important data, thus maintaining awareness without overwhelming the operator
Solution Approach 2:
The patent implements dynamics by making the user interface adaptive and responsive to operator needs. The system dynamically adjusts the level of detail and information presentation based on the current operational context, operator workload, and detected events, thereby maintaining optimal situational awareness and operator performance under varying conditions
Data Source
Figure 1~2
Figure 3
AI summary
A method and system for optimizing a human supervised cyber-physical system 10 determines a state of the human operator based on data from multiple psycho physiological sensors 40, determines a state of each of multiple cyber-physical systems 50 in the human supervised cyber-physical system based on data provided by the cyber-physical systems 50, and fuses the state of the human operator and the state of each of the plurality of cyber-physical systems 50 into a single state of the human supervised cyber-physical system. The single state is then used to generate recommendations for optimizing a user interface 30 and to generate high level control signals for the cyber-physical systems 50.