Intelligent Control Management of Autonomous Air Vehicles
Abstract
There are many issues in the general area of cooperative control of unmanned vehicles; one of particular interest is cooperative path planning and mission planning in a dynamic scenario with moving targets and moving obstacles. A dynamic scenario prevents usually the use of many algorithms due to their inherently high computational cost. The report briefly overviews some existing procedures used to solve both path planning and mission planning problems, and then proposes alternative algorithms which have a lower computational cost. In particular, we propose a path-planning procedure based on the Constrained Delaunay Triangulation, and the geometric properties of the in-centers of triangles. This procedure is not optimal from the analytical standpoint but it has several advantages for real-time applications because it allows slower sampling times and produces safer paths. The proposed path planning method takes into account areas of the scenario that may be more dangerous for the flight vehicle, by simply summing a term to the length of each sub-path depending of the dangerousness of the zone it crosses. The report presents also a sub-optimal mission planning algorithm based on a dynamic clustering of the targets in order to have a less myopic view of the entire scenario. The procedure is feasible in terms of total computational load, with respect to an optimal solution, which is known to be NP-hard and not achievable in polynomial time.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 2006
- Accession Number
- ADA463037
Entities
People
- Andrea Bracci
- Lorenzo Pollini
- Mario Innocenti
Organizations
- University of Pisa