Resilience and Guaranteed Task Completion for Partially Unknown Nonlinear Control Systems
Abstract
When performing in a contested environment, hostile actions and adverse events inevitably degrade the system s capabilities. In response, our work will enable provably correct control synthesis and task assignment for a system subject to a mid-mission adverse event. The project will result in theory and algorithms that (i) determine prior to the mission whether the system will be able to complete its task following the event, regardless of the event s exact nature, (ii) if the system is resilient, provide in real time a provably correct control signal that drives the system to complete the task following the event, and (iii) if the system is not resilient, propose in real time an alternative task that is guaranteed to be achievable given incomplete knowledge about new system dynamics, and design an appropriate control signal. Our running motivation is an aircraft experiencing partial loss of control authority, change in nonlinear and possibly timevarying dynamics, or degradation of actuator capabilities due to damage on some of the control surfaces or a remote adversarial takeover.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 29, 2024
- Source ID
- FA95502310131
Entities
People
- Melkior Ornik
Organizations
- Air Force Office of Scientific Research
- United States Air Force
- University of Illinois Urbana–Champaign