Trajectory optimization for constrained UAVs: A Virtual Target Vehicle approach

Abstract

In this paper we propose a novel approach for trajectory optimization for constrained Unmanned Aerial Vehicles (UAVs). With regard to the classical trajectory optimization problem, we take a Virtual Target Vehicle (VTV) perspective by introducing a virtual target that plays the role of an additional control input. Based on a nonlinear projection operator optimal control technique and extending the concepts of the maneuver regulation framework, we propose a trajectory optimization based strategy to compute, for any given desired path with a specified desired speed profile, the (local) optimal feasible trajectory that best approximates the desired one. The optimization procedure takes explicitly into account the extra flexibility of the VTV by changing (during the transient period) the velocity of the virtual target with the benefit of improving the convergence of the solver to obtain the optimal feasible path, and also avoid the singularities that occur in some maneuver regulation techniques described in the literature. We provide numerical computations for three testing scenarios that illustrates the effectiveness of the proposed strategy.