Online path optimization for unmanned aerial vehicles under steady wind

In this paper, online path optimization method for Unmanned Aerial Vehicles (UAV) by adjusting parameters of flight path is proposed. Especially, the energy efficient path for UAVs under the presence of steady wind is discussed. Although the relation between the energy consumption and the flight path parameters is hard to model explicitly in general, which induces the difficulty to adopt gradient based optimization techniques. On the other hand, actual energy consumption can be measured, e.g. by measuring fuel consumption rate, and it is natural to assume that the rate is a function of flight path. This paper proposes to perturb the flight path in order to estimate the gradient of the energy consumption rate that is used for the optimization, which is based on Simultaneous Perturbation Stochastic Approximation (SPSA) method. Numerical results are shown to evaluate the proposed approach, and the method succeeded to improve the energy efficiency.