PROCEEDINGS ARTICLE | May 23, 2011
Proc. SPIE. 8045, Unmanned Systems Technology XIII
KEYWORDS: Unmanned aerial vehicles, Genetic algorithms, Visualization, Sensors, Geographic information systems, Sensor performance, Genetics, Intelligence systems, Neodymium, Optimization (mathematics)
Unmanned aerial systems (UAS) have proven themselves to be indispensable in providing intelligence, surveillance, and
reconnaissance (ISR) over the battlefield. Constellations of heterogeneous, multi-purpose UAS are being tasked to
provide ISR in an unpredictable environment. This necessitates the dynamic replanning of critical missions as weather
conditions change, new observation targets are identified, aircraft are lost or equipment malfunctions, and new airspace
restrictions are introduced. We present a method to generate coordinated mission plans for constellations of UAS with
multiple flight goals and potentially competing objectives, and update them on demand as the operational situation
changes. We use a fast evolutionary algorithm-based, multi-objective optimization technique. The updated flight routes
maintain continuity by considering where the ISR assets have already flown and where they still need to go. Both the
initial planning and replanning take into account factors such as area of analysis coverage, restricted operating zones,
maximum control station range, adverse weather effects, military terrain value, and sensor performance. Our results
demonstrate that by constraining the space of potential solutions using an intelligently-formed air maneuver network
with a subset of potential airspace corridors and navigational waypoints, we can ensure global optimization for multiple
objectives considering the situation both before and after the replanning is initiated. We employ sophisticated
visualization techniques using a geographic information system to help the user 'look under the hood" of the algorithms
to understand the effectiveness and viability of the generated ISR mission plans and identify potential gaps in coverage.
