Next Generation Adaptive Cyber Physical Human Systems
Abstract
Cyber-Physical-Human Systems (CPHS) are a class of safety-critical, sociotechnical systems in which the interactions between the physical system and cyber elements that control its operation are influenced by human agents. The key distinguishing feature of CPHS are that human (agents) can intervene in cyber-physical system (CPS) task execution to redirect the system, or supply needed information, not just to assume full control, or exercise manual-override of CPS processes or tasks. CPHS typically allow other systems, devices, and data streams to connect/disconnect as needed during mission execution. In adaptive CPHS, humans collaborate with the cyber-physical elements to jointly accomplish tasks and adapt as needed to changing contexts to accomplish mission goals. The adaptation exploits prior knowledge and new knowledge gleaned from online machine learning while accommodating human cognitive constraints. A DoD-relevant problem scenario concerned with maintaining physical security of a stationary transport aircraft using building mounted cameras and a team of quadcopters that collectively maintain perimeter coverage is used to demonstrate the capabilities of the adaptive CPHS. The exemplar scenario is used to illustrate the interactions between the human and the cyber-physical system, as well as ongoing adaptations to restore loss of perimeter coverage arising from ongoing disruptions (e.g., loss of a quadcopter). This final report presents our accomplishments on this effort.
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 04, 2019
- Accession Number
- AD1099997
Entities
People
- Ayesha Madni
- Azad M. Madni
- Dan Erwin
- Edwin Ordoukhanian
- Parisa Pouya
- Shatad Purohit
Organizations
- Stevens Institute of Technology