UTN- FRC -Producción Académica de Investigación y Desarrollo

Permanent URI for this communityhttp://48.217.138.120/handle/20.500.12272/1932

Browse

Author: search.filters.author.Paz, ClaudioSubject: search.filters.subject.Quaternions

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A comparison of Bayesian filters for orientation estimation
    (XV Reunio´n de Trabajo en Procesamiento de la Informacion y Control, 2013) Paz, Claudio; Pérez Paina, Gonzalo; Toloza, Hugo
    This work presents an empirical study that compares the performance of Bayesian filters for orientation estimation using data provided by an Inertial Measurement Unit. A two-stages measure ment update was implemented for different variants of Kalman and particles filters using quaternions as orientation representation method. Empirical results show that all tested algorithms converge to the correct orientation even if the initial orientation is unknown
  • Thumbnail Image
    Item
    Trajectory planning for an unmanned quadrotor.
    (Universidad Tecnológica Nacional Regional Córdoba., 2016) Pucheta, Martín Alejo; Alberto, Nicolás; Paz, Claudio; Pérez Paina, Gonzalo
    . The quadrotor is an unmanned aerial vehicle (UAV) popular for its low-cost and broad range of application fields, inspection in harvesting environments, surveillance, cinematographic filmation, among others. Most of these applications are easily achieved if a set of predefined programmed trajecto ries or tasks are selected from a database of automatic tasks safety performed inside the limit cycle and compatible with actuators bounds. This paper presents two controllers that simultaneously coordinates the position and orientation motions of a quadrotor. The first controller is based on Euler angles and the second on quaternions. The equations of motion for the quadrotor are computed and simulated using Euler angles by means of a first-order reduction of the nonlinear Newton-Euler equations. This simulator of the plant is used to supply the observed state to simulate both controllers. The generation of trajec- tories with bounded derivatives on the velocities is used to provide continuous actions on the actuators and also prevent their saturation, enabling the minimization of trajectory tracking errors and reducing or avoiding the overshooting. Finally, the simulation for three kinematic tasks, a circle, a 8-shape, and a square trajectory are shown to test the controller behaviors and adjust their gains.