System Identification of a Multi-Rotor UAV Actuator

摘要

The most widely used actuator system for small unmanned aerial vehicles, from fixed wing to rotary wing, is composed of a Brushless DC motor (BLDC), an Electronic Speed Controller (ESC), and a propeller. The complex electrical system, composed of the BLDC motor and ESC, drives its mechanical response. The goal of this work is to experimentally identify a simple model of the mechanical dynamics that captures and can predict the behavior of such a system. The first step towards realizing this goal and the subject of this paper is the identification of the actuator system in hover or zero forward velocity conditions. The structure of the dynamic model is determined from first principle modeling, static characterization, and regressor statistical contributions to the accuracy of the model. Results show that local behavior of the mechanical dynamics can be modeled by a first order linear system. The global behavior of the system is well described by a set of local first order linear models. In the flight operating range of the actuator, the first order model has a varying input gain but a relatively constant pole.