Optimizing time-varying performance and mission aborting policy in resource constrained missions

Abstract

Intensive efforts have been devoted to mission aborting systems. However, the existing models mostly assumed static performance or failed to consider limited resources (e.g., energy, budget). Motivated by practical applications like unmanned aerial vehicles (UAV), this paper relaxes those assumptions by modeling a resource-constrained system that must complete a required amount of work for a successful mission and accomplish further a return/rescue phase (RP) to survive the system. The operation phase (OP) of the mission may be aborted depending on the number of external shocks (e.g., electromagnetic interferences, radiations) the system has survived and the operation time elapsed, followed by a RP to save the asset. Probabilistic methods are proposed to evaluate the mission success probability (MSP) and system survival probability (SSP). An optimization problem is formulated and solved, which determines the joint optimal time-varying performance policy and OP aborting policy, maximizing the MSP while providing a required level of SSP. A case study of UAV executing a reconnaissance mission is carried out to demonstrate the suggested model and examine influence of the SSP level as well as shock parameters on the mission performance and optimal policies. The advantage of time-varying performance in enhancing the MSP is also demonstrated.