Trajectory optimization of an autonomous dynamic soaring UAV

Abstract

This paper has a single goal of finding admissible state trajectories for an unmanned aerial vehicle while utilizing dynamic soaring technique in a linear wind gradient. Loiter pattern trajectory of the small UAV is optimized by employing three non-linear optimal control problems. The whole atmospheric altitude is divided into three regions(below 100m, 100m–115m and above 115m) and each are assigned with carefully chosen wind velocities. The resultant optimized trajectories reveal the possibility of applying dynamic soaring to UAVs which take off with a particular initial velocity from any of the three altitude regions. Performance measures adopted are, maximizing specific energy and specific energy rate extracted by the vehicle, and minimizing the control effort delivered. A 3-DOF point mass model system dynamics of UAV is studied, and control variables are identified as bank angle and lift co-efficient. The state trajectories, euler angle variations, aerodynamic performances are analyzed by changing controls. A suitable wind profile is chosen. The state equations, controls, terminal conditions, objective function, and initial trajectory(initial conditions) are then incorporated into a MATLAB dynamic optimization routine using numerical methods and optimized with the help of genetic algorithm in order to produce converged optimal dynamic soaring trajectories. The results of this study could have a direct impact on the application of UAVs, both military and civilian, as a thrust-off situation could help in saving of fuel and thereby prolong duration of flight, extension of range, and endurance.