Showing papers by "Yasmina Bestaoui published in 2011"

Route Finding for An Autonomous Aircraft

Abstract: For mission safety and efficiency, the launching system has to find an appropriate route to reach the mission goal in the presence of environment requirements and system constraints. The A* planning algorithm proposes an optimal solution to find this route taking into account the path cost and system constraints. Positions and shapes of known obstacles are introduced in the application of the A* algorithm in 3D space. Some simulation results are presented.

Analysis of Time Optimal 3D Paths for an Autonomous Aircraft with a Piecewise Constant Acceleration

Abstract: Trajectory generation is a fundamental part of planning for an autonomous aerial vehicle. For the purpose of flight path generation, it is usually sufficient to treat only the translational motion. One component of the weather that greatly affects an aircraft trajectory is the wind. Study of the accessibility of this nonlinear affine system with drift makes use of the Lie algebra rank condition. The second part of this paper presents 3D time optimal translational trajectories characterization for an aircraft in steady wind. If unaccounted for, winds can substantially degrade the performance of an autonomous aircraft guidance system. We consider finding a time optimal trajectory for an airplane from some starting point and orientation to some final point and orientation, assuming that the system has independent bounded control over the acceleration as well as the turning rates for the flight path and heading angles. Through the use of the Pontryagin maximum principle, we characterize the time optimal trajectories for the system.

Trajectory generation in a 3D flight plan with obstacle avoidance for an airborne launch craft

Abstract: For mission safety and efficiency, the airborne launch vehicle has to find an appropriate route to reach the mission goal satisfying some environment requirements and system constraints. A modified 3D waypoints generation path based on an improved version of the A* algorithm is proposed to find an optimal flight plan solution. This simple geometric path planning procedure can be implemented in real time in order to plan a new reference trajectory. In a presence of detected obstacles such as turbulence zones, no-fly zones, storms, etc., the information about the environment is regularly updated. This route reaches one or multiple goal points important for the mission success in order of their priority. The improved A* capabilities are tested via simulations in different scenario.

3D curves with a prescribed curvature and torsion for an aerial robot

Abstract: The objective of this paper is to generate a desired flight path to be followed by an aerial robot. A curve with discontinuous curvature and torsion is not appropriate for smooth motions for any vehicle architecture. The flyable path should be smooth, i.e. without twists and cusps. The smoothness of the path is determined by amount of bending of the path, measured by curvature and torsion of the path. Three different classes of curves are presented. First, constant curvature and torsion followed by a linear variation versus the curvilinear abscissa then a quadratic variation. Finally, the problem of manoeuvre between two trim helices of different curvature and torsion is tackled.