Micro air vehicle trajectory generation in pitch plane

Abstract

For autonomous operations, unmanned micro air vehicles depend on novel trajectory generation schemes. Trajectories parallel to the ground are fairly well understood in surveillance and reconnaissance contexts. In addition, trajectories with altitude as one of the navigational parameters are also envisioned for aerial robots, urban terrain coverage, etc. In these missions, trajectory generation in pitch plane becomes an important problem. In this article, a linear algebraic procedure is applied to generate the pitch plane trajectories. That is, when a terrain (latitude, longitude and altitude) is sensed from a global positioning system, a trajectory to navigate the aircraft along the terrain is presented. By linear algebra, the state vector is spanned using a set of known vectors and some of their scalars (coefficients of spanning vectors) are solved in an optimization framework so that the state variables mapping the desired terrain are determined. After the trajectories are generated, normally smoothening is recommended so that the trajectories are continuous for autonomous control development. Thus, an extended Kalman filter is applied to smooth the trajectories. A nonlinear micro air vehicle model is considered to illustrate the trajectory generation and smoothening scheme.