@InProceedings{MedeirosRodKerSanShi:2015:InTaAs,
author = "Medeiros, Ivo Paix{\~a}o de and Rodrigues, Leonardo Ramos and
Kern, Christian Stromtmann and Santos, Rafael Duarte Coelho dos
and Shiguemori, Elcio",
affiliation = "Embraer and Embraer and Embraer and {Instituto Nacional de
Pesquisas Espaciais (INPE)}",
title = "Integrated Task Assignment and Maintenance Recommendation based on
System Architecture and PHM Information for UAVs",
year = "2015",
organization = "Annual IEEE International Systems Conference, 9.",
abstract = "Recently, Unmanned Aerial Vehicle (UAV) have become important in
supporting military and civilian organizations for gathering
information and assessing situations from remote locations. As
such, squad of UAVs have been deployed to gather data from
different locations, thus providing operators with the most
up-to-date situational information for the areas they cover. Based
on that evidence, there is no doubt that these vehicles have been
providing helpful capability, but abnormal conditions and degraded
system components can render their capabilities ineffective.
Prognostics and Health Monitoring (PHM) has the potential to
extend a vehicle lifecycle, as an enabler that provides, in
advance, health information regards to critical systems for the
operation of an UAV. PHM (Prognostics and Health Monitoring) can
be defined as the capability of assessing the health state,
predicting impending failures and forecasting the expected RUL
(Remaining Useful Life) of a component based on a set of
measurements collected from systems. Bearing that in mind, another
important concept that could be built upon PHM is IVHM (Integrate
Vehicle Health Management); that is the unified capability of
integrating PHM within a framework of available resources and
operational demand. Whenever, it is applied to decision making
problems relating to flight operations decision and maintenance
optimization; such as task assignment and maintenance
recommendation, that one comes into play as a powerful decision
support tool to provide operational efficiency, low ownership
costs and high availability. This big picture is the IVHM
proposal, which could for example integrates PHM information
relating to critical components, UAV systems architecture, mission
time constraints, tasks priority for a given mission and safety
concerns in order to assign UAVs to their waypoint in a sortie and
to plan maintenance aiming to perform a higher dispatchability;
integrating available resources and operational demand scenario.
So that, this work aims to integrate task assignment and
maintenance recommendation, both based on PHM information, for
UAVs (Unmanned Aerial Vehicle) Swarm. Task assignment is the
problem of assigning a vehicle to a task and to perform this. This
paper uses a PHM-based assignment solution; this solution takes
into account mission time, task priority and vehicle health state.
Another concept in this paper is maintenance recommendation, that
is the operation of defining which component should receive
maintenance action. It is computed by an algorithm that takes into
account PHM information, system architecture and safety margins.
As mentioned earlier, both take advantage of a combination of PHM
information and system architecture to compute UAV health state,
so called here S-RUL (System Level Remaining Useful Life); S-RUL
is the integration element in this solution granting that the
squad is assigned to their task considering the chance they could
come into failure and that they are receiving proper maintenance
considering the vehicle architecture or mission demand, such way
the squad have high readiness. In S-RUL, the decision maker does
not have to deal with a set of component level RULs. Instead, the
S-RUL provides information related to the time when the whole
system will stop working or the mission constraints do not match.
In the case study, a simplified pitch control system is used to
illustrate the application of the proposed method to UAVs Swarm.",
conference-location = "Vancouver, Canada",
conference-year = "13-16 apr.",
language = "en",
urlaccessdate = "28 abr. 2024"
}