Authors: A. Baudkoubeh; M. Farrokhi
Addresses: Department of Electrical Engineering, Iran University of Science and Technology, Tehran, 16846, Iran. ' Department of Electrical Engineering, Iran University of Science and Technology, Tehran, 16846, Iran
Abstract: This paper utilises a class of mesh adaptive direct search method to design an optimal path for unmanned aerial vehicles (UAVs). To this end, a multi-objective optimisation problem is considered for simultaneous optimisation of some conflicting objective functions under different kinds of vehicle and mission constraints. Since the path planning for UAVs in a large geographical area is a typical large-scale optimisation problem, to avoid memory and computational intensive issues, different techniques such as constructing an adaptive mesh, polling, and barrier approach are incorporated in the proposed algorithm. The proposed method is tested under different scenarios and various realistic terrain environments. The results show effectiveness of the proposed method in guiding UAVs to the final destination by providing near-optimal feasible paths quickly and effectively. The results will also be compared with the genetic algorithm approach, which has been recently used for path planning.
Keywords: nonlinear optimisation; optimal path planning; mesh adaptive direct search; MADS; unmanned aerial vehicles; UAVs; genetic algorithms.
International Journal of Mathematical Modelling and Numerical Optimisation, 2011 Vol.2 No.4, pp.422 - 440
Received: 15 Jan 2011
Accepted: 24 Jun 2011
Published online: 26 Mar 2015 *