Implementing dubins airplane paths on fixed-wing uavs

Abstract

A well-known path-planning technique for mobile robots or planar aerial vehicles is to use Dubins paths, which are minimum-distance paths between two configurations subject to the constraints of the Dubins car model. An extension of this method to a three-dimensional Dubins airplane model has recently been proposed. This chapter builds on that work showing a complete architecture for implementing Dubins airplane paths on small fixed-wing UAVs. The existing Dubins airplane model is modified to be more consistent with the kinematics of a fixed-wing aircraft. The chapter then shows how a recently proposed vector-field method can be used to design a guidance law that causes the Dubins airplane model to follow straight-line and helical paths. Dubins airplane paths are more complicated than Dubins car paths because of the altitude component. Based on the difference between the altitude of the start and end configurations, Dubins airplane paths can be classified as low, medium, or high altitude gain. While for medium and high altitude gain there are many different Dubins airplane paths, this chapter proposes selecting the path that maximizes the average altitude throughout the maneuver. The proposed architecture is implemented on a six degree-of-freedom Matlab/Simulink simulation of an Aerosonde UAV, and results from this simulation demonstrate the effectiveness of the technique.